system

Abstract

We provide the system. [Solution] A device for collecting document data including job postings, A device that applies language processing to the aforementioned document data and detects fraudulent features, A device for evaluating the trustworthiness of information providers, A device that identifies and deletes or reports job postings with fraudulent characteristics, A device that sends a warning to the user's terminal regarding fraudulent job postings, A system that includes this.

Description

Technical Field

[0001] The technology of the present disclosure relates to a system.

Background Art

[0002] Patent Document 1 discloses a persona chatbot control method performed by at least one processor, including steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In recent years, the number of fraudulent job offers through SNS and job hunting sites has been increasing rapidly, and their methods have also become sophisticated. Therefore, it is difficult to quickly detect and respond to these fraudulent job offers. As a result, the reliability of SNS operators is damaged, and the risk that users will be involved in fraudulent job offers is also increasing. There is a need for a system that can quickly and accurately detect and respond to fraudulent job offers.

Means for Solving the Problems

[0005] This invention provides a system that collects text data containing job postings using a specific algorithm and detects fraudulent features by applying natural language processing. Furthermore, it has the function of identifying potentially fraudulent job postings by evaluating their reliability using the poster's past activity history and user feedback, and then deleting them or reporting them to law enforcement. This makes it possible to quickly identify fraudulent job postings and prevent crimes before they occur.

[0006] "Job postings" refer to information that companies and organizations use when recruiting new personnel, and include details such as job description, salary conditions, and application procedures.

[0007] "Text data" refers to a collection of information composed of strings of characters, such as job postings, that can be read by electronic devices.

[0008] "Fraudulent characteristics" refer to features that indicate illegality or inaccuracies compared to typical job postings, and which have the potential to cause social problems.

[0009] "Natural language processing" is a technology that understands and processes human language, and is used by computers to analyze language data and grasp its meaning.

[0010] "Trustworthiness" is a numerical value or indicator that evaluates whether specific information or its provider is trustworthy, and is based on past activity history and user feedback.

[0011] An "algorithm" is a set of steps or procedures performed in a specific order to achieve a particular objective.

[0012] "Reporting" refers to the act of reporting fraudulent or illegal activities to a specific agency, with the aim of raising awareness of the problem and encouraging appropriate action. [Brief explanation of the drawing]

[0013] [Figure 1]This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of the data processing device and smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Figure 11] This is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] This is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] This is a sequence diagram showing the processing flow of the data processing system in Example 2, when an emotion engine is combined. [Figure 14] This is a sequence diagram showing the processing flow of the data processing system in Application Example 2, which combines an emotion engine. [Modes for carrying out the invention]

[0014] Hereinafter, an example of an embodiment of the system relating to the technology of this disclosure will be described with reference to the attached drawings.

[0015] First, the terms used in the following description will be explained.

[0016] In the following embodiments, a labeled processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0017] In the following embodiments, a labeled RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0018] In the following embodiments, a labeled storage is one or more non-volatile storage devices that store various programs and various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes, and the like.

[0019] In the following embodiments, a labeled communication I / F (Interface) is an interface including a communication processor and an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark), and the like.

[0020] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0021] [First Embodiment]

[0022] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0023] As shown in Figure 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0024] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0025] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0026] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0027] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0028] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0029] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0030] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0031] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0032] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0033] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0034] This invention provides a system for effectively and quickly detecting fraudulent job postings and taking appropriate action based on that information. The operation of the system will be described below as an embodiment.

[0035] The server first collects text data, including job postings, from social media and job search websites. This data is collected periodically using APIs and scraping techniques, so new posts are incorporated into the system almost in real time.

[0036] Next, the server applies natural language processing techniques to the collected text data to analyze posts that may have fraudulent features. This process uses a pre-trained model to detect specific keywords, phrases, and their context. For example, the terminal will be wary of posts containing phrases commonly used in fraudulent job postings, such as "high income" and "immediate payment."

[0037] The server also evaluates the trustworthiness of the poster. This evaluation uses the poster's past activity history and user feedback. For example, a poster who has frequently made fraudulent posts in the past will be given a low trustworthiness rating. This information is used to further narrow down the possibility of fraudulent job postings.

[0038] For posts deemed to be fraudulent job postings, the server will either automatically delete them or request confirmation from the social media administrator. Furthermore, it will report them to law enforcement agencies as necessary. To prevent users from applying to fraudulent job postings, their devices will display warning messages and take appropriate measures.

[0039] For example, if a job posting containing phrases like "No experience necessary! Immediate payment, high income" is posted on a social networking site, the server will immediately detect this information and determine that it is highly likely to be fraudulent. Subsequently, the post will be promptly deleted, and the necessary information will be provided to law enforcement. During this process, a warning will be displayed on the user's device, prompting them to take action to maintain their safety.

[0040] Such systems enable the rapid detection and response to fraudulent job postings, thereby improving the security of the internet environment.

[0041] The following describes the processing flow.

[0042] Step 1:

[0043] The server collects job information as text data from social media and job search websites. This collection is performed continuously in real time using APIs and web scraping techniques.

[0044] Step 2:

[0045] The server begins preprocessing the collected text data. This process removes unnecessary HTML tags and advertisements, generating pure text data optimized for analysis.

[0046] Step 3:

[0047] The server uses a natural language processing engine to analyze text data and extract specific keywords and phrases. For example, it checks whether terms like "high income" or "same-day payment" are included.

[0048] Step 4:

[0049] The server performs contextual analysis to understand the overall tone and intent of the job posting. This involves using natural language processing algorithms to assess the potential for fraud based on the flow and structure of the sentences.

[0050] Step 5:

[0051] The server evaluates the trustworthiness of the poster. Based on past posting history and user feedback, it calculates a trustworthiness score and pays particular attention to posts with low scores.

[0052] Step 6:

[0053] The server triggers an automated deletion process based on information it identifies as fraudulent job postings. If necessary, it sends a notification to the social media administrator requesting their confirmation.

[0054] Step 7:

[0055] If the server determines that it is necessary to notify law enforcement, it will initiate an automated notification process and provide the necessary information.

[0056] Step 8:

[0057] The device displays a warning message to the user regarding fraudulent job postings. The user can review this warning and receive instructions on how to take safety precautions.

[0058] Step 9:

[0059] The server saves the results of the above processing to a database and uses them for future analysis and improvement. This forms a feedback loop to improve the accuracy of the model.

[0060] (Example 1)

[0061] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0062] The increase in fraudulent information online, particularly in job postings, is raising the risk of users becoming victims of scams and unethical recruitment practices. Such fraudulent information can lead to the leakage of users' personal information and financial losses, making it crucial to efficiently and quickly detect such information and implement appropriate countermeasures.

[0063] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0064] In this invention, the server includes means for collecting information from an information network using data communication technology, means for analyzing fraudulent features by applying natural language processing technology, and means for evaluating the trustworthiness of information providers and generating an evaluation score based on that trustworthiness. This enables the rapid detection of fraudulent information, appropriate deletion or notification, and warning display to users, thereby providing safe and reliable information.

[0065] "Data communication technology" refers to technologies that enable the transmission and reception of data over information networks.

[0066] An "information network" is a network that allows computers to exchange data with each other, including the internet and LANs.

[0067] "Natural language processing technology" is a technology that enables computers to understand and analyze human language.

[0068] "Fraudulent features" refer to specific patterns or linguistic elements that may indicate fraudulent information.

[0069] An "information provider" refers to an individual or organization that disseminates information on an information network.

[0070] A "trust score" is a quantitative measure of an information provider's reliability, calculated based on their past behavior and evaluations.

[0071] "Means for generating evaluation scores" refers to algorithms and processes for quantifying the reliability of information providers.

[0072] "Deletion or notification" refers to actions that remove malicious information from a network or alert those involved.

[0073] A "warning message" is a message displayed on a user's device that informs them that certain information may be fraudulent.

[0074] This invention is a system aimed at quickly and effectively detecting fraudulent job postings on an information network and taking appropriate measures. Specific embodiments of the system are described below.

[0075] The server first uses APIs and scraping techniques to collect text data, including job postings, from social media and job search websites. This collected data is updated in near real-time, and new information is incorporated into the system.

[0076] Next, the server applies natural language processing techniques to the collected text data. In this step, a generative AI model is used to detect keywords and phrases with specific malicious characteristics. For example, it pays attention to posts that contain phrases such as "high income" or "instant payment."

[0077] Furthermore, the server evaluates the trustworthiness of the information provider. This evaluation takes into account the information provider's past activity history and user feedback. Based on the evaluation results, a trustworthiness score is calculated for the information provider and used to detect fraudulent information.

[0078] If information is deemed fraudulent, the server will automatically delete it or notify relevant authorities. Additionally, the terminal will display a warning message to the user, prompting them to take action to ensure their security.

[0079] As a concrete example, consider a scenario where a user finds a job posting on social media with content such as "No experience necessary! Immediate payment, high income." The server immediately identifies this information as fraudulent and deletes it, and a warning message appears on the user's device stating, "This job posting may be fraudulent. Please be careful."

[0080] This invention makes it possible to quickly detect and respond to fraudulent job postings on the internet, thereby protecting the safety of users.

[0081] Examples of prompt messages are as follows:

[0082] "Please explain in detail how to detect social media posts containing fraudulent job postings in real time."

[0083] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0084] Step 1:

[0085] Data collection

[0086] The server uses APIs or scraping techniques to collect text data containing job postings from social media and job sites. The input to this process is a specified URL or hashtag from a designated information network, and the output is a list of the collected text data. For example, the server periodically calls the Twitter API to retrieve posts containing the "job postings" tag and saves them to a database.

[0087] Step 2:

[0088] Text analysis

[0089] The server applies natural language processing techniques to the collected text data. The input for this step is the text data obtained in the previous step, and the output is the analysis result indicating whether or not it is potentially fraudulent. By using a generative AI model to detect specific keywords and phrases, the characteristics of fraudulent job postings are identified. For example, posts containing expressions such as "high income" and "immediate payment" are targeted for attention.

[0090] Step 3:

[0091] Confidence Rating

[0092] The server evaluates the trustworthiness of information providers who post text data. The input for this process is data about the poster's past activity history and user feedback, and the output is a trustworthiness score. If a provider has a history of making fraudulent posts, the server assigns them a low trustworthiness score.

[0093] Step 4:

[0094] Identifying and responding to fraudulent information

[0095] The server identifies job postings deemed fraudulent and takes appropriate action. The inputs for this step are analysis results and confidence scores, while the outputs are deletion orders and notification orders. The server either automatically deletes the fraudulent information or initiates a protocol to notify the administrator. It also notifies law enforcement agencies where appropriate.

[0096] Step 5:

[0097] User notifications

[0098] The terminal displays a warning message to the user. The input to this process is a warning notification sent from the server, and the output is a warning message displayed on the screen. A message such as "This job posting may be fraudulent. Please be careful." is displayed to prevent the user from accidentally applying for fraudulent information.

[0099] (Application Example 1)

[0100] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0101] Online platforms frequently receive fraudulent job postings, leading to users mistakenly applying to these jobs. This increases the risk of users becoming victims of fraud or suffering other disadvantages. Furthermore, it is difficult to quickly identify and address fraudulent job postings, highlighting the need for effective measures to ensure user safety.

[0102] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0103] In this invention, the server includes means for collecting document data containing job postings, means for applying language processing to the document data to detect fraudulent features, and means for evaluating the trustworthiness of the information provider. This enables the rapid detection of fraudulent job postings and the issuance of warnings to the user's terminal, thereby reducing the risk of users becoming involved in fraudulent job postings and allowing them to conduct job-seeking activities safely.

[0104] "Job postings" refer to information regarding job openings and employment conditions.

[0105] "Document data" refers to all electronically stored information, including characters, symbols, and descriptions.

[0106] "Language processing" refers to the technology that enables computers to understand, analyze, and process natural language.

[0107] "Fraudulent features" refer to characteristics of information that are judged to contain intent to commit fraud or misuse.

[0108] An "information provider" refers to an entity that provides or publishes information online.

[0109] "Trustworthiness" refers to a measure used to evaluate the reliability and credibility of information providers and their information.

[0110] A "reliability score" refers to a numerical representation of trustworthiness calculated based on the information provider's past actions and feedback.

[0111] To realize this invention, a system is built through the cooperation of a server and a user's terminal. The server first collects document data, including job postings. This is typically done using APIs or scraping techniques. Next, the server applies natural language processing techniques to the collected document data to detect fraudulent features. A natural language processing library (e.g., spaCy) is used for this language processing. To identify information with fraudulent features, generative AI models using TENSORFLOW® or PyTorch are utilized.

[0112] Furthermore, the server evaluates the trustworthiness of information providers and generates a trustworthiness score based on past activity history and user feedback. Based on this evaluation criterion, fraudulent job postings can be quickly identified. Job postings deemed fraudulent are automatically deleted or reported by the server. In addition, warnings are displayed on the user's device in real time, prompting them to take measures to prevent damage from fraudulent job postings.

[0113] For example, if a job posting such as "No experience necessary, immediate payment, high income" is displayed on a device, the server immediately analyzes the content and, if it determines that there is a possibility of fraud, displays a warning to the user saying, "Suspicious job posting has been detected. Please investigate carefully before applying."

[0114] Examples of prompts for a generative AI model:

[0115] "Please evaluate the likelihood that the following job posting is fraudulent: 'No experience necessary! Immediate payment, high income.'"

[0116] This system allows users to safely conduct job-seeking activities online.

[0117] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0118] Step 1:

[0119] The server collects document data, including job postings, from online platforms using APIs or scraping techniques. Inputs include URLs and access information for each platform, while output is job postings stored as text data. This data is then stored in a database for subsequent analysis.

[0120] Step 2:

[0121] The server performs vocabulary analysis on the collected text data using natural language processing libraries such as spaCy. The input is document data containing job postings, and the output is a list of words and phrases, as well as the results of contextual analysis. Here, the frequency of occurrence of specific keywords and phrases is calculated, and features that may indicate fraud are extracted.

[0122] Step 3:

[0123] The server uses a generative AI model based on TensorFlow or PyTorch to analyze fraudulent features based on the results of natural language processing. The input is the feature data obtained in step 2, and the output is a score or flag indicating the likelihood of fraudulent job postings. This score quantifies the degree of suspicion of fraud and is used in the next evaluation stage.

[0124] Step 4:

[0125] The server evaluates the trustworthiness of information providers using their past activity history and user feedback. The input is the information provider's account data and past posting history, and the output is a trustworthiness score. This score is used to determine the threshold for identifying fraudulent job postings.

[0126] Step 5:

[0127] The server notifies the user's terminal in real time of information identified as fraudulent job postings and displays a warning message. The input is the ID of the job posting identified as fraudulent in the previous stage and the user's identification information, and the output is a warning message displayed on the terminal screen. This operation allows users to recognize the risk before applying for fraudulent job postings.

[0128] Step 6:

[0129] The server removes fraudulent job postings and, if necessary, reports them to administrators or law enforcement. Input is information identified as fraudulent, and output is the removal of that information or a report of the fraudulent activity. This helps maintain a safer online environment.

[0130] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0131] This invention provides a system for more effectively detecting fraudulent job postings, combining natural language processing, poster trustworthiness evaluation, and an emotion engine that recognizes user sentiment. The operation of the system will be specifically described below as an embodiment.

[0132] First, the server collects job information from social media and job search websites. This data is dynamically retrieved using APIs and scraping techniques, and any new posts are processed immediately.

[0133] The server applies natural language processing to the collected data to extract fraudulent features. This process identifies specific keywords and contexts, revealing patterns unique to fraudulent job postings. For example, the terminal identifies posts containing risky expressions such as "mass recruitment" and "immediate start."

[0134] Furthermore, the server analyzes the poster's past activity and calculates a trust score. This score is adjusted based on user feedback and the poster's history. Posts from users with low trust scores are reviewed with particular care.

[0135] A key feature of this system is its integrated emotion engine. The emotion engine recognizes the user's emotions, and the device provides appropriate warning messages and guidance. This process customizes the warning content according to the user's emotional state, designed to reduce anxiety and apprehension. For example, if the user expresses anxiety, the device will present more detailed and reassuring guidance.

[0136] As a concrete example, if a job posting contains phrases such as "high income" or "no experience necessary," and shows signs of fraud, the server will immediately detect this information and observe the decrease in its trustworthiness. The terminal will monitor the user's reaction and display an appropriate warning through the sentiment engine. This example allows users to be appropriately vigilant against fraudulent job postings and use the information with peace of mind.

[0137] This system provides an effective means to improve the security of job postings and enhance the user experience.

[0138] The following describes the processing flow.

[0139] Step 1:

[0140] The server collects job information from social media and job sites using APIs and web scraping. This ensures that the latest information is incorporated into the system in real time.

[0141] Step 2:

[0142] The server preprocesses the collected text data, removing unnecessary information to prepare it for analysis. This allows the natural language processing engine to process the data in an optimized manner.

[0143] Step 3:

[0144] The server applies natural language processing to analyze keywords and phrases with fraudulent characteristics. This involves using pre-trained algorithms to identify common patterns in fraudulent job postings.

[0145] Step 4:

[0146] The server calculates a trust score based on the poster's activity history and user feedback. This score assesses the poster's trustworthiness and helps in detecting fraudulent activity.

[0147] Step 5:

[0148] If any fraudulent characteristics are detected, the server identifies the job posting as fraudulent and automatically notifies the administrator to request its deletion or correction.

[0149] Step 6:

[0150] The server activates an emotion engine to analyze the user's emotions. This helps the server understand the user's feelings towards fraudulent job postings and consider appropriate countermeasures.

[0151] Step 7:

[0152] The device displays warning messages and suggested actions to the user based on analysis results from the emotion engine. The content of the guidance automatically changes according to the user's emotional state.

[0153] Step 8:

[0154] Users receive the displayed warnings and take appropriate action as needed. This helps maintain a safe internet environment.

[0155] Step 9:

[0156] The server records all processing results in a database, forming a feedback loop that is used to improve future models and increase detection accuracy.

[0157] (Example 2)

[0158] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0159] In recent years, the number of job postings on the internet has increased, but some of these include malicious and fraudulent job postings, posing a risk to users who mistakenly act based on this information. Furthermore, traditional systems are insufficient in detecting fraudulent information, making it difficult for users to use job postings safely. In addition, users must judge for themselves whether a job posting is fraudulent, which often causes psychological burden. Under these circumstances, there is a need for a system that improves the security of job postings and allows users to use information with peace of mind.

[0160] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0161] In this invention, the server includes means for collecting text information including job postings, means for applying natural language processing to detect fraudulent features, means for evaluating the trustworthiness of the poster, and means for recognizing the user's emotions and providing a corresponding warning message. This makes it possible to quickly evaluate the trustworthiness of job postings and provide appropriate warnings to users. As a result, users can maintain appropriate vigilance against fraudulent job postings and use the information safely.

[0162] "Job postings" refer to information about occupations and jobs that is made public to those who wish to find employment or change jobs.

[0163] "Text information" refers to digital data composed of characters, specifically data stored in the form of text or documents.

[0164] "Natural language processing" refers to the technology that enables computers to understand and analyze human language.

[0165] "Fraudulent features" refer to suspicious or questionable patterns or keywords that are not typically found in job postings.

[0166] "Contributor reliability" refers to an indicator that quantifies the reliability of the person providing the job posting based on their past behavior and reputation.

[0167] "Means of recognizing emotions" refers to the technologies and processes used to identify and respond to human emotions.

[0168] A "warning message" refers to a notification or message displayed to alert the user.

[0169] "Means of deletion or reporting" refers to processes or functions for removing inappropriate information from a system or notifying relevant authorities of inappropriate activity.

[0170] This invention is a system for ensuring the security of job postings and is implemented using a server and terminals. First, the server collects job postings from various sources on the internet. These sources include social networking services (SNS) and job posting websites. The collected information is obtained using an API or automatically collected from web pages using scraping technology.

[0171] Next, the server applies natural language processing to the collected text information to identify fraudulent features. Natural language processing uses algorithms for specific keywords and contextual analysis. This detects patterns indicating fraudulent job postings, thereby identifying fraudulent job listings.

[0172] Furthermore, the server evaluates the trustworthiness of the job posting poster. Trustworthiness is calculated based on past activity history and user feedback. Posts with low trustworthiness receive special attention and are subject to further review.

[0173] Meanwhile, the device uses emotion recognition technology to monitor user reactions and provides appropriate warning messages to the user through an emotion engine. For example, if a user expresses anxiety about a job posting, the device displays reassuring information. This allows the user to use the information with confidence.

[0174] As a concrete example of this system, let's consider a scenario where the server detects job postings containing distinctive phrases such as "no experience necessary" and "high pay." If the reliability of these job postings is low, the terminal uses an emotion engine to issue a warning to the user. The user can then review this warning and make a safer choice.

[0175] Examples of prompt statements to input into a generative AI model are as follows:

[0176] "Please explain in detail the methods used to detect fraudulent job postings using AI. Clearly state the specific algorithms and technologies used, and also touch upon the system design that takes user sentiment into consideration."

[0177] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0178] Step 1:

[0179] The server collects job postings from internet sources. Specifically, it retrieves data using APIs or employs web scraping techniques. Inputs include specified URLs or API endpoints, and the retrieved job postings are output. This information is stored in a database and used for subsequent processing.

[0180] Step 2:

[0181] The server applies natural language processing to the collected job posting text data. In this process, the collected text data is used as input, and an algorithm is used to analyze keywords and context. As part of the data processing, specific keywords and phrases are extracted, and job postings with fraudulent characteristics are output. For example, risky terms such as "mass recruitment" and "no experience necessary" are detected.

[0182] Step 3:

[0183] The server evaluates the trustworthiness of job posting posters. Input includes the poster's past activity history and user feedback. Based on this, a trustworthiness score is calculated and the result is output. During data processing, an evaluation algorithm is applied using this historical information to adjust the score. For posters with low trustworthiness, job postings are reviewed more carefully.

[0184] Step 4:

[0185] The device recognizes the user's emotions and generates warning messages using an emotion engine. The input is user response data, which is passed through an emotion recognition algorithm. This analyzes the user's emotional state and outputs a corresponding warning message based on the results. For example, if the user shows signs of anxiety, information to reassure them will be presented.

[0186] Step 5:

[0187] The server identifies job postings with fraudulent characteristics and deletes or reports them as necessary. Here, the output data from step 2 is used as input, and a process is executed to filter out information deemed fraudulent. Finally, details of deleted data and reports are output, enabling action to be taken against fraudulent job postings.

[0188] Through these steps, the system enhances the reliability of job postings and provides an environment where users can confidently conduct their job search.

[0189] (Application Example 2)

[0190] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0191] In recent years, with the proliferation of job postings on the internet, the number of victims of false or fraudulent information has increased. Such fraudulent information can pose a significant risk to job seekers, making it necessary to provide effective means of detection and warning. However, conventional methods have the challenge of not being able to detect fraudulent information in real time and provide appropriate warnings. Furthermore, there is a lack of means to provide warnings tailored to the individual circumstances and emotions of job seekers.

[0192] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0193] In this invention, the server includes means for collecting text data containing job postings, means for applying natural language processing to the text data to detect fraudulent features, and means for evaluating the trustworthiness of the poster. This enables the detection of fraudulent job postings in real time and the provision of visual and auditory warnings to users. Furthermore, by using emotion recognition technology, it is possible to customize and provide appropriate warning content according to the user's emotions, creating an environment in which job postings can be used with peace of mind.

[0194] "Job postings" refer to information provided to job seekers online or in print media, including details such as job description, salary conditions, and work location.

[0195] "Text data" refers to a collection of information composed of characters, including textual data such as job postings.

[0196] "Natural language processing" is a technology that enables computers to understand and analyze human language, and is a method for extracting useful information from text data.

[0197] "Fraudulent characteristics" refer to the properties of information that may be false or misleading, and indicate fraudulent job postings through specific keywords or patterns.

[0198] "Means" refer to the methods or devices used to achieve a specific objective, and are elements that constitute a system.

[0199] "Poster's credibility" is an evaluation of the reliability of the person providing the job posting, and is an indicator based on past activity history and feedback.

[0200] A "warning message" is a warning statement displayed to inform a user that there is potentially malicious information, and is provided through visual or auditory means.

[0201] "User" refers to an individual or legal entity that views or uses job postings through this system.

[0202] "Emotional state" refers to the psychological reactions and moods that users exhibit when they receive information.

[0203] "Optimization" is the process of adjusting or improving something to its most effective or efficient state according to specific criteria.

[0204] To implement this invention, the server first collects job information publicly available on the internet. This involves using technologies that efficiently collect the latest job information from job sites and social media using web crawlers and APIs. The collected text data is analyzed by natural language processing software to detect information with fraudulent characteristics. Here, natural language processing technologies such as Amazon Comprehend can be used.

[0205] Furthermore, the server assigns a trustworthiness rating to each poster, based on their past posting history and user ratings stored in the database. Trustworthiness analysis is efficiently performed by calculating the rating score using machine learning platforms such as Amazon SageMaker.

[0206] The primary device used by users is smart glasses, which display real-time warning messages based on collected and analyzed job information via an augmented reality display. Voice feedback using Amazon Polly is also considered, designed to appropriately capture the user's attention.

[0207] For example, when a user views a job posting containing phrases like "high pay, no experience necessary," the system instantly assesses the reliability of the information and displays a warning message on the glasses such as "This job posting carries risks." Voice guidance allows users to select job postings with confidence. In this way, the system provides an innovative service that improves the user experience through fraud detection of job postings.

[0208] Examples of prompt statements to input into a generative AI model include the following:

[0209] "Try developing an application for smart glasses that analyzes new job postings, identifies signs of fraud, and warns users. How can you ensure a safe user experience as a result?"

[0210] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0211] Step 1:

[0212] The server collects job information from the internet. During collection, it uses web crawlers and APIs to retrieve the latest job postings from job sites and social media. The input is the URL or API endpoint of each job posting, and the output is the collected raw job text data. The server then organizes this data for subsequent processing steps.

[0213] Step 2:

[0214] The server applies natural language processing to the collected job posting text data. This process uses natural language processing tools such as Amazon Comprehend to extract keywords and contexts from the text data to detect fraudulent features. The input is the raw job posting text data, and the output is the text portions identified as potentially fraudulent based on the analyzed information. The server records this as a list of fraudulent features.

[0215] Step 3:

[0216] The server evaluates the trustworthiness of job posters based on the job information it has already analyzed. The server references historical data and user feedback, and generates a trustworthiness score using tools such as Amazon SageMaker. The input consists of the analyzed job information and historical database data, and the output is a numerical trustworthiness score. This trustworthiness score will be used for future warning decisions.

[0217] Step 4:

[0218] The device provides users with visual and audible warning messages. It displays warning messages in real time via an augmented reality display and provides audio feedback using Amazon Polly. Inputs are a list of fraudulent features and confidence scores, while outputs are warning messages and audio guidance for the user. This allows the device to help users feel confident receiving information even when fraudulent activity is detected.

[0219] Step 5:

[0220] The system evaluates job postings based on information provided by the user. Users refer to warning messages from their device to avoid potentially fraudulent information. The input is warning information provided by the device, and the output is the user's safe judgment. This system allows users to use job postings with greater peace of mind.

[0221] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0222] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0223] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0224] [Second Embodiment]

[0225] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0226] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0227] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0228] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0229] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0230] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0231] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0232] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0233] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0234] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0235] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0236] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0237] This invention provides a system for effectively and quickly detecting fraudulent job postings and taking appropriate action based on that information. The operation of the system will be described below as an embodiment.

[0238] The server first collects text data, including job postings, from social media and job search websites. This data is collected periodically using APIs and scraping techniques, so new posts are incorporated into the system almost in real time.

[0239] Next, the server applies natural language processing techniques to the collected text data to analyze posts that may have fraudulent features. This process uses a pre-trained model to detect specific keywords, phrases, and their context. For example, the terminal will be wary of posts containing phrases commonly used in fraudulent job postings, such as "high income" and "immediate payment."

[0240] The server also evaluates the trustworthiness of the poster. This evaluation uses the poster's past activity history and user feedback. For example, a poster who has frequently made fraudulent posts in the past will be given a low trustworthiness rating. This information is used to further narrow down the possibility of fraudulent job postings.

[0241] For posts deemed to be fraudulent job postings, the server will either automatically delete them or request confirmation from the social media administrator. Furthermore, it will report them to law enforcement agencies as necessary. To prevent users from applying to fraudulent job postings, their devices will display warning messages and take appropriate measures.

[0242] For example, if a job posting containing phrases like "No experience necessary! Immediate payment, high income" is posted on a social networking site, the server will immediately detect this information and determine that it is highly likely to be fraudulent. Subsequently, the post will be promptly deleted, and the necessary information will be provided to law enforcement. During this process, a warning will be displayed on the user's device, prompting them to take action to maintain their safety.

[0243] Such systems enable the rapid detection and response to fraudulent job postings, thereby improving the security of the internet environment.

[0244] The following describes the processing flow.

[0245] Step 1:

[0246] The server collects job information as text data from social media and job search websites. This collection is performed continuously in real time using APIs and web scraping techniques.

[0247] Step 2:

[0248] The server begins preprocessing the collected text data. This process removes unnecessary HTML tags and advertisements, generating pure text data optimized for analysis.

[0249] Step 3:

[0250] The server uses a natural language processing engine to analyze text data and extract specific keywords and phrases. For example, it checks whether terms like "high income" or "same-day payment" are included.

[0251] Step 4:

[0252] The server performs contextual analysis to understand the overall tone and intent of the job posting. This involves using natural language processing algorithms to assess the potential for fraud based on the flow and structure of the sentences.

[0253] Step 5:

[0254] The server evaluates the trustworthiness of the poster. Based on past posting history and user feedback, it calculates a trustworthiness score and pays particular attention to posts with low scores.

[0255] Step 6:

[0256] The server triggers an automated deletion process based on information it identifies as fraudulent job postings. If necessary, it sends a notification to the social media administrator requesting their confirmation.

[0257] Step 7:

[0258] If the server determines that it is necessary to notify law enforcement, it will initiate an automated notification process and provide the necessary information.

[0259] Step 8:

[0260] The device displays a warning message to the user regarding fraudulent job postings. The user can review this warning and receive instructions on how to take safety precautions.

[0261] Step 9:

[0262] The server saves the results of the above processing to a database and uses them for future analysis and improvement. This forms a feedback loop to improve the accuracy of the model.

[0263] (Example 1)

[0264] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0265] The increase in fraudulent information online, particularly in job postings, is raising the risk of users becoming victims of scams and unethical recruitment practices. Such fraudulent information can lead to the leakage of users' personal information and financial losses, making it crucial to efficiently and quickly detect such information and implement appropriate countermeasures.

[0266] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0267] In this invention, the server includes means for collecting information from an information network using data communication technology, means for analyzing fraudulent features by applying natural language processing technology, and means for evaluating the trustworthiness of information providers and generating an evaluation score based on that trustworthiness. This enables the rapid detection of fraudulent information, appropriate deletion or notification, and warning display to users, thereby providing safe and reliable information.

[0268] "Data communication technology" refers to technologies that enable the transmission and reception of data over information networks.

[0269] An "information network" is a network that allows computers to exchange data with each other, including the internet and LANs.

[0270] "Natural language processing technology" is a technology that enables computers to understand and analyze human language.

[0271] "Fraudulent features" refer to specific patterns or linguistic elements that may indicate fraudulent information.

[0272] An "information provider" refers to an individual or organization that disseminates information on an information network.

[0273] A "trust score" is a quantitative measure of an information provider's reliability, calculated based on their past behavior and evaluations.

[0274] "Means for generating evaluation scores" refers to algorithms and processes for quantifying the reliability of information providers.

[0275] "Deletion or notification" refers to actions that remove malicious information from a network or alert those involved.

[0276] A "warning message" is a message displayed on a user's device that informs them that certain information may be fraudulent.

[0277] This invention is a system aimed at quickly and effectively detecting fraudulent job postings on an information network and taking appropriate measures. Specific embodiments of the system are described below.

[0278] The server first uses APIs and scraping techniques to collect text data, including job postings, from social media and job search websites. This collected data is updated in near real-time, and new information is incorporated into the system.

[0279] Next, the server applies natural language processing technology to the collected text data. In this step, a generative AI model is used to detect keywords and phrases with specific fraudulent characteristics. For example, attention is paid to posts containing expressions such as "high income" and "same-day payment".

[0280] Furthermore, the server evaluates the reliability of the information provider. This evaluation takes into account the information provider's past activity history and feedback from users. Based on the evaluation results, a reliability score for the information provider is calculated and used for the detection of fraudulent information.

[0281] For information determined to be fraudulent, the server automatically deletes it or notifies the relevant authorities. In addition, the terminal displays a warning message to the user and prompts actions to ensure safety.

[0282] As a specific example, consider the case where a user finds a job offer on SNS with content such as "No experience OK! Same-day payment, high income". The server immediately determines this information to be fraudulent and deletes it, and a warning "This job offer may be fraudulent. Please be careful." is displayed on the terminal.

[0283] With this invention, it becomes possible to quickly detect and respond to fraudulent job offers on the Internet, protecting the safety of users.

[0284] Examples of prompt texts are as follows.

[0285] "Please explain in detail the method of detecting SNS posts containing fraudulent job offers in real time."

[0286] The flow of specific processing in Example 1 will be described using FIG. 11.

[0287] Step 1:

[0288] Data collection

[0289] The server uses APIs or scraping techniques to collect text data containing job postings from social media and job sites. The input to this process is a specified URL or hashtag from a designated information network, and the output is a list of the collected text data. For example, the server periodically calls the Twitter API to retrieve posts containing the "job postings" tag and saves them to a database.

[0290] Step 2:

[0291] Text analysis

[0292] The server applies natural language processing techniques to the collected text data. The input for this step is the text data obtained in the previous step, and the output is the analysis result indicating whether or not it is potentially fraudulent. By using a generative AI model to detect specific keywords and phrases, the characteristics of fraudulent job postings are identified. For example, posts containing expressions such as "high income" and "immediate payment" are targeted for attention.

[0293] Step 3:

[0294] Confidence Rating

[0295] The server evaluates the trustworthiness of information providers who post text data. The input for this process is data about the poster's past activity history and user feedback, and the output is a trustworthiness score. If a provider has a history of making fraudulent posts, the server assigns them a low trustworthiness score.

[0296] Step 4:

[0297] Identifying and responding to fraudulent information

[0298] The server identifies job postings deemed fraudulent and takes appropriate action. The inputs for this step are analysis results and confidence scores, while the outputs are deletion orders and notification orders. The server either automatically deletes the fraudulent information or initiates a protocol to notify the administrator. It also notifies law enforcement agencies where appropriate.

[0299] Step 5:

[0300] User notifications

[0301] The terminal displays a warning message to the user. The input to this process is a warning notification sent from the server, and the output is a warning message displayed on the screen. A message such as "This job posting may be fraudulent. Please be careful." is displayed to prevent the user from accidentally applying for fraudulent information.

[0302] (Application Example 1)

[0303] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0304] Online platforms frequently receive fraudulent job postings, leading to users mistakenly applying to these jobs. This increases the risk of users becoming victims of fraud or suffering other disadvantages. Furthermore, it is difficult to quickly identify and address fraudulent job postings, highlighting the need for effective measures to ensure user safety.

[0305] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0306] In this invention, the server includes means for collecting document data containing job offer information, means for applying language processing to the document data to detect irregular features, and means for evaluating the reliability of the information sender. As a result, it becomes possible to quickly detect fraudulent job offer information, send a warning to the user's terminal, reduce the risk that the user will be involved in fraudulent job offers, and enable safe job hunting activities.

[0307] "Job offer information" refers to information regarding job recruitment and employment conditions.

[0308] "Document data" refers to all electronically stored information including characters, symbols, and descriptions.

[0309] "Language processing" refers to the technology by which a computer understands, analyzes, and processes natural language.

[0310] "Irregular features" refer to the characteristics of information that are judged to include intentions for fraud or abuse.

[0311] "Information sender" refers to the entity that provides or discloses information online.

[0312] "Reliability" refers to a measure for evaluating the reliability and credibility of an information sender or information.

[0313] "Reliability score" refers to the numerical representation of the reliability calculated based on the past behavior and feedback of the information sender.

[0314] To implement this invention, a server and a user's terminal cooperate to build a system. First, the server collects document data containing job offer information. For this, methods using APIs and scraping technologies are common. Next, the server applies language processing technology to the collected document data to detect irregular features. For this language processing, a natural language processing library (e.g., spaCy) is used. To identify information with irregular features, a generative AI model using TensorFlow or PyTorch is utilized.

[0315] Furthermore, the server evaluates the trustworthiness of information providers and generates a trustworthiness score based on past activity history and user feedback. Based on this evaluation criterion, fraudulent job postings can be quickly identified. Job postings deemed fraudulent are automatically deleted or reported by the server. In addition, warnings are displayed on the user's device in real time, prompting them to take measures to prevent damage from fraudulent job postings.

[0316] For example, if a job posting such as "No experience necessary, immediate payment, high income" is displayed on a device, the server immediately analyzes the content and, if it determines that there is a possibility of fraud, displays a warning to the user saying, "Suspicious job posting has been detected. Please investigate carefully before applying."

[0317] Examples of prompts for a generative AI model:

[0318] "Please evaluate the likelihood that the following job posting is fraudulent: 'No experience necessary! Immediate payment, high income.'"

[0319] This system allows users to safely conduct job-seeking activities online.

[0320] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0321] Step 1:

[0322] The server collects document data, including job postings, from online platforms using APIs or scraping techniques. Inputs include URLs and access information for each platform, while output is job postings stored as text data. This data is then stored in a database for subsequent analysis.

[0323] Step 2:

[0324] The server performs vocabulary analysis on the collected text data using natural language processing libraries such as spaCy. The input is document data containing job postings, and the output is a list of words and phrases, as well as the results of contextual analysis. Here, the frequency of occurrence of specific keywords and phrases is calculated, and features that may indicate fraud are extracted.

[0325] Step 3:

[0326] The server uses a generative AI model based on TensorFlow or PyTorch to analyze fraudulent features based on the results of natural language processing. The input is the feature data obtained in step 2, and the output is a score or flag indicating the likelihood of fraudulent job postings. This score quantifies the degree of suspicion of fraud and is used in the next evaluation stage.

[0327] Step 4:

[0328] The server evaluates the trustworthiness of information providers using their past activity history and user feedback. The input is the information provider's account data and past posting history, and the output is a trustworthiness score. This score is used to determine the threshold for identifying fraudulent job postings.

[0329] Step 5:

[0330] The server notifies the user's terminal in real time of information identified as fraudulent job postings and displays a warning message. The input is the ID of the job posting identified as fraudulent in the previous stage and the user's identification information, and the output is a warning message displayed on the terminal screen. This operation allows users to recognize the risk before applying for fraudulent job postings.

[0331] Step 6:

[0332] The server removes fraudulent job postings and, if necessary, reports them to administrators or law enforcement. Input is information identified as fraudulent, and output is the removal of that information or a report of the fraudulent activity. This helps maintain a safer online environment.

[0333] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0334] This invention provides a system for more effectively detecting fraudulent job postings, combining natural language processing, poster trustworthiness evaluation, and an emotion engine that recognizes user sentiment. The operation of the system will be specifically described below as an embodiment.

[0335] First, the server collects job information from social media and job search websites. This data is dynamically retrieved using APIs and scraping techniques, and any new posts are processed immediately.

[0336] The server applies natural language processing to the collected data to extract fraudulent features. This process identifies specific keywords and contexts, revealing patterns unique to fraudulent job postings. For example, the terminal identifies posts containing risky expressions such as "mass recruitment" and "immediate start."

[0337] Furthermore, the server analyzes the poster's past activity and calculates a trust score. This score is adjusted based on user feedback and the poster's history. Posts from users with low trust scores are reviewed with particular care.

[0338] A key feature of this system is its integrated emotion engine. The emotion engine recognizes the user's emotions, and the device provides appropriate warning messages and guidance. This process customizes the warning content according to the user's emotional state, designed to reduce anxiety and apprehension. For example, if the user expresses anxiety, the device will present more detailed and reassuring guidance.

[0339] As a concrete example, if a job posting contains phrases such as "high income" or "no experience necessary," and shows signs of fraud, the server will immediately detect this information and observe the decrease in its trustworthiness. The terminal will monitor the user's reaction and display an appropriate warning through the sentiment engine. This example allows users to be appropriately vigilant against fraudulent job postings and use the information with peace of mind.

[0340] This system provides an effective means to improve the security of job postings and enhance the user experience.

[0341] The following describes the processing flow.

[0342] Step 1:

[0343] The server collects job information from social media and job sites using APIs and web scraping. This ensures that the latest information is incorporated into the system in real time.

[0344] Step 2:

[0345] The server preprocesses the collected text data, removing unnecessary information to prepare it for analysis. This allows the natural language processing engine to process the data in an optimized manner.

[0346] Step 3:

[0347] The server applies natural language processing to analyze keywords and phrases with fraudulent characteristics. This involves using pre-trained algorithms to identify common patterns in fraudulent job postings.

[0348] Step 4:

[0349] The server calculates a trust score based on the poster's activity history and user feedback. This score assesses the poster's trustworthiness and helps in detecting fraudulent activity.

[0350] Step 5:

[0351] If any fraudulent characteristics are detected, the server identifies the job posting as fraudulent and automatically notifies the administrator to request its deletion or correction.

[0352] Step 6:

[0353] The server activates an emotion engine to analyze the user's emotions. This helps the server understand the user's feelings towards fraudulent job postings and consider appropriate countermeasures.

[0354] Step 7:

[0355] The device displays warning messages and suggested actions to the user based on analysis results from the emotion engine. The content of the guidance automatically changes according to the user's emotional state.

[0356] Step 8:

[0357] Users receive the displayed warnings and take appropriate action as needed. This helps maintain a safe internet environment.

[0358] Step 9:

[0359] The server records all processing results in a database, forming a feedback loop that is used to improve future models and increase detection accuracy.

[0360] (Example 2)

[0361] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0362] In recent years, the number of job postings on the internet has increased, but some of these include malicious and fraudulent job postings, posing a risk to users who mistakenly act based on this information. Furthermore, traditional systems are insufficient in detecting fraudulent information, making it difficult for users to use job postings safely. In addition, users must judge for themselves whether a job posting is fraudulent, which often causes psychological burden. Under these circumstances, there is a need for a system that improves the security of job postings and allows users to use information with peace of mind.

[0363] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0364] In this invention, the server includes means for collecting text information including job postings, means for applying natural language processing to detect fraudulent features, means for evaluating the trustworthiness of the poster, and means for recognizing the user's emotions and providing a corresponding warning message. This makes it possible to quickly evaluate the trustworthiness of job postings and provide appropriate warnings to users. As a result, users can maintain appropriate vigilance against fraudulent job postings and use the information safely.

[0365] "Job postings" refer to information about occupations and jobs that is made public to those who wish to find employment or change jobs.

[0366] "Text information" refers to digital data composed of characters, specifically data stored in the form of text or documents.

[0367] "Natural language processing" refers to the technology that enables computers to understand and analyze human language.

[0368] "Fraudulent features" refer to suspicious or questionable patterns or keywords that are not typically found in job postings.

[0369] "Contributor reliability" refers to an indicator that quantifies the reliability of the person providing the job posting based on their past behavior and reputation.

[0370] "Means of recognizing emotions" refers to the technologies and processes used to identify and respond to human emotions.

[0371] A "warning message" refers to a notification or message displayed to alert the user.

[0372] "Means of deletion or reporting" refers to processes or functions for removing inappropriate information from a system or notifying relevant authorities of inappropriate activity.

[0373] This invention is a system for ensuring the security of job postings and is implemented using a server and terminals. First, the server collects job postings from various sources on the internet. These sources include social networking services (SNS) and job posting websites. The collected information is obtained using an API or automatically collected from web pages using scraping technology.

[0374] Next, the server applies natural language processing to the collected text information to identify fraudulent features. Natural language processing uses algorithms for specific keywords and contextual analysis. This detects patterns indicating fraudulent job postings, thereby identifying fraudulent job listings.

[0375] Furthermore, the server evaluates the trustworthiness of the job posting poster. Trustworthiness is calculated based on past activity history and user feedback. Posts with low trustworthiness receive special attention and are subject to further review.

[0376] Meanwhile, the device uses emotion recognition technology to monitor user reactions and provides appropriate warning messages to the user through an emotion engine. For example, if a user expresses anxiety about a job posting, the device displays reassuring information. This allows the user to use the information with confidence.

[0377] As a concrete example of this system, let's consider a scenario where the server detects job postings containing distinctive phrases such as "no experience necessary" and "high pay." If the reliability of these job postings is low, the terminal uses an emotion engine to issue a warning to the user. The user can then review this warning and make a safer choice.

[0378] Examples of prompt statements to input into a generative AI model are as follows:

[0379] "Please explain in detail the methods used to detect fraudulent job postings using AI. Clearly state the specific algorithms and technologies used, and also touch upon the system design that takes user sentiment into consideration."

[0380] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0381] Step 1:

[0382] The server collects job postings from internet sources. Specifically, it retrieves data using APIs or employs web scraping techniques. Inputs include specified URLs or API endpoints, and the retrieved job postings are output. This information is stored in a database and used for subsequent processing.

[0383] Step 2:

[0384] The server applies natural language processing to the collected job posting text data. In this process, the collected text data is used as input, and an algorithm is used to analyze keywords and context. As part of the data processing, specific keywords and phrases are extracted, and job postings with fraudulent characteristics are output. For example, risky terms such as "mass recruitment" and "no experience necessary" are detected.

[0385] Step 3:

[0386] The server evaluates the trustworthiness of job posting posters. Input includes the poster's past activity history and user feedback. Based on this, a trustworthiness score is calculated and the result is output. During data processing, an evaluation algorithm is applied using this historical information to adjust the score. For posters with low trustworthiness, job postings are reviewed more carefully.

[0387] Step 4:

[0388] The device recognizes the user's emotions and generates warning messages using an emotion engine. The input is user response data, which is passed through an emotion recognition algorithm. This analyzes the user's emotional state and outputs a corresponding warning message based on the results. For example, if the user shows signs of anxiety, information to reassure them will be presented.

[0389] Step 5:

[0390] The server identifies job postings with fraudulent characteristics and deletes or reports them as necessary. Here, the output data from step 2 is used as input, and a process is executed to filter out information deemed fraudulent. Finally, details of deleted data and reports are output, enabling action to be taken against fraudulent job postings.

[0391] Through these steps, the system enhances the reliability of job postings and provides an environment where users can confidently conduct their job search.

[0392] (Application Example 2)

[0393] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the smart glasses 214 as the "terminal".

[0394] In recent years, with the proliferation of job postings on the internet, the number of victims of false or fraudulent information has increased. Such fraudulent information can pose a significant risk to job seekers, making it necessary to provide effective means of detection and warning. However, conventional methods have the challenge of not being able to detect fraudulent information in real time and provide appropriate warnings. Furthermore, there is a lack of means to provide warnings tailored to the individual circumstances and emotions of job seekers.

[0395] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0396] In this invention, the server includes means for collecting text data containing job postings, means for applying natural language processing to the text data to detect fraudulent features, and means for evaluating the trustworthiness of the poster. This enables the detection of fraudulent job postings in real time and the provision of visual and auditory warnings to users. Furthermore, by using emotion recognition technology, it is possible to customize and provide appropriate warning content according to the user's emotions, creating an environment in which job postings can be used with peace of mind.

[0397] "Job postings" refer to information provided to job seekers online or in print media, including details such as job description, salary conditions, and work location.

[0398] "Text data" refers to a collection of information composed of characters, including textual data such as job postings.

[0399] "Natural language processing" is a technology that enables computers to understand and analyze human language, and is a method for extracting useful information from text data.

[0400] "Fraudulent characteristics" refer to the properties of information that may be false or misleading, and indicate fraudulent job postings through specific keywords or patterns.

[0401] "Means" refer to the methods or devices used to achieve a specific objective, and are elements that constitute a system.

[0402] "Poster's credibility" is an evaluation of the reliability of the person providing the job posting, and is an indicator based on past activity history and feedback.

[0403] A "warning message" is a warning statement displayed to inform a user that there is potentially malicious information, and is provided through visual or auditory means.

[0404] "User" refers to an individual or legal entity that views or uses job postings through this system.

[0405] "Emotional state" refers to the psychological reactions and moods that users exhibit when they receive information.

[0406] "Optimization" is the process of adjusting or improving something to its most effective or efficient state according to specific criteria.

[0407] To implement this invention, the server first collects job information publicly available on the internet. This involves using technologies that efficiently collect the latest job information from job sites and social media using web crawlers and APIs. The collected text data is analyzed by natural language processing software to detect information with fraudulent characteristics. Here, natural language processing technologies such as Amazon Comprehend can be used.

[0408] Furthermore, the server assigns a trustworthiness rating to each poster, based on their past posting history and user ratings stored in the database. Trustworthiness analysis is efficiently performed by calculating the rating score using machine learning platforms such as Amazon SageMaker.

[0409] The primary device used by users is smart glasses, which display real-time warning messages based on collected and analyzed job information via an augmented reality display. Voice feedback using Amazon Polly is also considered, designed to appropriately capture the user's attention.

[0410] For example, when a user views a job posting containing phrases like "high pay, no experience necessary," the system instantly assesses the reliability of the information and displays a warning message on the glasses such as "This job posting carries risks." Voice guidance allows users to select job postings with confidence. In this way, the system provides an innovative service that improves the user experience through fraud detection of job postings.

[0411] Examples of prompt statements to input into a generative AI model include the following:

[0412] "Try developing an application for smart glasses that analyzes new job postings, identifies signs of fraud, and warns users. How can you ensure a safe user experience as a result?"

[0413] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0414] Step 1:

[0415] The server collects job information from the internet. During collection, it uses web crawlers and APIs to retrieve the latest job postings from job sites and social media. The input is the URL or API endpoint of each job posting, and the output is the collected raw job text data. The server then organizes this data for subsequent processing steps.

[0416] Step 2:

[0417] The server applies natural language processing to the collected job posting text data. This process uses natural language processing tools such as Amazon Comprehend to extract keywords and contexts from the text data to detect fraudulent features. The input is the raw job posting text data, and the output is the text portions identified as potentially fraudulent based on the analyzed information. The server records this as a list of fraudulent features.

[0418] Step 3:

[0419] The server evaluates the trustworthiness of job posters based on the job information it has already analyzed. The server references historical data and user feedback, and generates a trustworthiness score using tools such as Amazon SageMaker. The input consists of the analyzed job information and historical database data, and the output is a numerical trustworthiness score. This trustworthiness score will be used for future warning decisions.

[0420] Step 4:

[0421] The device provides users with visual and audible warning messages. It displays warning messages in real time via an augmented reality display and provides audio feedback using Amazon Polly. Inputs are a list of fraudulent features and confidence scores, while outputs are warning messages and audio guidance for the user. This allows the device to help users feel confident receiving information even when fraudulent activity is detected.

[0422] Step 5:

[0423] The system evaluates job postings based on information provided by the user. Users refer to warning messages from their device to avoid potentially fraudulent information. The input is warning information provided by the device, and the output is the user's safe judgment. This system allows users to use job postings with greater peace of mind.

[0424] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0425] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0426] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0427] [Third Embodiment]

[0428] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0429] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0430] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0431] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0432] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0433] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0434] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0435] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0436] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0437] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0438] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0439] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0440] This invention provides a system for effectively and quickly detecting fraudulent job postings and taking appropriate action based on that information. The operation of the system will be described below as an embodiment.

[0441] The server first collects text data, including job postings, from social media and job search websites. This data is collected periodically using APIs and scraping techniques, so new posts are incorporated into the system almost in real time.

[0442] Next, the server applies natural language processing techniques to the collected text data to analyze posts that may have fraudulent features. This process uses a pre-trained model to detect specific keywords, phrases, and their context. For example, the terminal will be wary of posts containing phrases commonly used in fraudulent job postings, such as "high income" and "immediate payment."

[0443] The server also evaluates the trustworthiness of the poster. This evaluation uses the poster's past activity history and user feedback. For example, a poster who has frequently made fraudulent posts in the past will be given a low trustworthiness rating. This information is used to further narrow down the possibility of fraudulent job postings.

[0444] For posts deemed to be fraudulent job postings, the server will either automatically delete them or request confirmation from the social media administrator. Furthermore, it will report them to law enforcement agencies as necessary. To prevent users from applying to fraudulent job postings, their devices will display warning messages and take appropriate measures.

[0445] For example, if a job posting containing phrases like "No experience necessary! Immediate payment, high income" is posted on a social networking site, the server will immediately detect this information and determine that it is highly likely to be fraudulent. Subsequently, the post will be promptly deleted, and the necessary information will be provided to law enforcement. During this process, a warning will be displayed on the user's device, prompting them to take action to maintain their safety.

[0446] Such systems enable the rapid detection and response to fraudulent job postings, thereby improving the security of the internet environment.

[0447] The following describes the processing flow.

[0448] Step 1:

[0449] The server collects job information as text data from social media and job search websites. This collection is performed continuously in real time using APIs and web scraping techniques.

[0450] Step 2:

[0451] The server begins preprocessing the collected text data. This process removes unnecessary HTML tags and advertisements, generating pure text data optimized for analysis.

[0452] Step 3:

[0453] The server uses a natural language processing engine to analyze text data and extract specific keywords and phrases. For example, it checks whether terms like "high income" or "same-day payment" are included.

[0454] Step 4:

[0455] The server performs contextual analysis to understand the overall tone and intent of the job posting. This involves using natural language processing algorithms to assess the potential for fraud based on the flow and structure of the sentences.

[0456] Step 5:

[0457] The server evaluates the trustworthiness of the poster. Based on past posting history and user feedback, it calculates a trustworthiness score and pays particular attention to posts with low scores.

[0458] Step 6:

[0459] The server triggers an automated deletion process based on information it identifies as fraudulent job postings. If necessary, it sends a notification to the social media administrator requesting their confirmation.

[0460] Step 7:

[0461] If the server determines that it is necessary to notify law enforcement, it will initiate an automated notification process and provide the necessary information.

[0462] Step 8:

[0463] The device displays a warning message to the user regarding fraudulent job postings. The user can review this warning and receive instructions on how to take safety precautions.

[0464] Step 9:

[0465] The server saves the results of the above processing to a database and uses them for future analysis and improvement. This forms a feedback loop to improve the accuracy of the model.

[0466] (Example 1)

[0467] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0468] The increase in fraudulent information online, particularly in job postings, is raising the risk of users becoming victims of scams and unethical recruitment practices. Such fraudulent information can lead to the leakage of users' personal information and financial losses, making it crucial to efficiently and quickly detect such information and implement appropriate countermeasures.

[0469] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0470] In this invention, the server includes means for collecting information from an information network using data communication technology, means for analyzing fraudulent features by applying natural language processing technology, and means for evaluating the trustworthiness of information providers and generating an evaluation score based on that trustworthiness. This enables the rapid detection of fraudulent information, appropriate deletion or notification, and warning display to users, thereby providing safe and reliable information.

[0471] "Data communication technology" refers to technologies that enable the transmission and reception of data over information networks.

[0472] An "information network" is a network that allows computers to exchange data with each other, including the internet and LANs.

[0473] "Natural language processing technology" is a technology that enables computers to understand and analyze human language.

[0474] "Fraudulent features" refer to specific patterns or linguistic elements that may indicate fraudulent information.

[0475] An "information provider" refers to an individual or organization that disseminates information on an information network.

[0476] A "trust score" is a quantitative measure of an information provider's reliability, calculated based on their past behavior and evaluations.

[0477] "Means for generating evaluation scores" refers to algorithms and processes for quantifying the reliability of information providers.

[0478] "Deletion or notification" refers to actions that remove malicious information from a network or alert those involved.

[0479] A "warning message" is a message displayed on a user's device that informs them that certain information may be fraudulent.

[0480] This invention is a system aimed at quickly and effectively detecting fraudulent job postings on an information network and taking appropriate measures. Specific embodiments of the system are described below.

[0481] The server first uses APIs and scraping techniques to collect text data, including job postings, from social media and job search websites. This collected data is updated in near real-time, and new information is incorporated into the system.

[0482] Next, the server applies natural language processing techniques to the collected text data. In this step, a generative AI model is used to detect keywords and phrases with specific malicious characteristics. For example, it pays attention to posts that contain phrases such as "high income" or "instant payment."

[0483] Furthermore, the server evaluates the trustworthiness of the information provider. This evaluation takes into account the information provider's past activity history and user feedback. Based on the evaluation results, a trustworthiness score is calculated for the information provider and used to detect fraudulent information.

[0484] If information is deemed fraudulent, the server will automatically delete it or notify relevant authorities. Additionally, the terminal will display a warning message to the user, prompting them to take action to ensure their security.

[0485] As a concrete example, consider a scenario where a user finds a job posting on social media with content such as "No experience necessary! Immediate payment, high income." The server immediately identifies this information as fraudulent and deletes it, and a warning message appears on the user's device stating, "This job posting may be fraudulent. Please be careful."

[0486] This invention makes it possible to quickly detect and respond to fraudulent job postings on the internet, thereby protecting the safety of users.

[0487] Examples of prompt messages are as follows:

[0488] "Please explain in detail how to detect social media posts containing fraudulent job postings in real time."

[0489] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0490] Step 1:

[0491] Data collection

[0492] The server uses APIs or scraping techniques to collect text data containing job postings from social media and job sites. The input to this process is a specified URL or hashtag from a designated information network, and the output is a list of the collected text data. For example, the server periodically calls the Twitter API to retrieve posts containing the "job postings" tag and saves them to a database.

[0493] Step 2:

[0494] Text analysis

[0495] The server applies natural language processing techniques to the collected text data. The input for this step is the text data obtained in the previous step, and the output is the analysis result indicating whether or not it is potentially fraudulent. By using a generative AI model to detect specific keywords and phrases, the characteristics of fraudulent job postings are identified. For example, posts containing expressions such as "high income" and "immediate payment" are targeted for attention.

[0496] Step 3:

[0497] Confidence Rating

[0498] The server evaluates the trustworthiness of information providers who post text data. The input for this process is data about the poster's past activity history and user feedback, and the output is a trustworthiness score. If a provider has a history of making fraudulent posts, the server assigns them a low trustworthiness score.

[0499] Step 4:

[0500] Identifying and responding to fraudulent information

[0501] The server identifies job postings deemed fraudulent and takes appropriate action. The inputs for this step are analysis results and confidence scores, while the outputs are deletion orders and notification orders. The server either automatically deletes the fraudulent information or initiates a protocol to notify the administrator. It also notifies law enforcement agencies where appropriate.

[0502] Step 5:

[0503] User notifications

[0504] The terminal displays a warning message to the user. The input to this process is a warning notification sent from the server, and the output is a warning message displayed on the screen. A message such as "This job posting may be fraudulent. Please be careful." is displayed to prevent the user from accidentally applying for fraudulent information.

[0505] (Application Example 1)

[0506] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0507] Online platforms frequently receive fraudulent job postings, leading to users mistakenly applying to these jobs. This increases the risk of users becoming victims of fraud or suffering other disadvantages. Furthermore, it is difficult to quickly identify and address fraudulent job postings, highlighting the need for effective measures to ensure user safety.

[0508] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0509] In this invention, the server includes means for collecting document data containing job postings, means for applying language processing to the document data to detect fraudulent features, and means for evaluating the trustworthiness of the information provider. This enables the rapid detection of fraudulent job postings and the issuance of warnings to the user's terminal, thereby reducing the risk of users becoming involved in fraudulent job postings and allowing them to conduct job-seeking activities safely.

[0510] "Job postings" refer to information regarding job openings and employment conditions.

[0511] "Document data" refers to all electronically stored information, including characters, symbols, and descriptions.

[0512] "Language processing" refers to the technology that enables computers to understand, analyze, and process natural language.

[0513] "Fraudulent features" refer to characteristics of information that are judged to contain intent to commit fraud or misuse.

[0514] An "information provider" refers to an entity that provides or publishes information online.

[0515] "Trustworthiness" refers to a measure used to evaluate the reliability and credibility of information providers and their information.

[0516] A "reliability score" refers to a numerical representation of trustworthiness calculated based on the information provider's past actions and feedback.

[0517] To realize this invention, a system is built through the cooperation of a server and a user's terminal. The server first collects document data, including job postings. This is typically done using APIs or scraping techniques. Next, the server applies natural language processing techniques to the collected document data to detect fraudulent features. A natural language processing library (e.g., spaCy) is used for this language processing. To identify information with fraudulent features, generative AI models using TensorFlow or PyTorch are utilized.

[0518] Furthermore, the server evaluates the trustworthiness of information providers and generates a trustworthiness score based on past activity history and user feedback. Based on this evaluation criterion, fraudulent job postings can be quickly identified. Job postings deemed fraudulent are automatically deleted or reported by the server. In addition, warnings are displayed on the user's device in real time, prompting them to take measures to prevent damage from fraudulent job postings.

[0519] For example, if a job posting such as "No experience necessary, immediate payment, high income" is displayed on a device, the server immediately analyzes the content and, if it determines that there is a possibility of fraud, displays a warning to the user saying, "Suspicious job posting has been detected. Please investigate carefully before applying."

[0520] Examples of prompts for a generative AI model:

[0521] "Please evaluate the likelihood that the following job posting is fraudulent: 'No experience necessary! Immediate payment, high income.'"

[0522] This system allows users to safely conduct job-seeking activities online.

[0523] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0524] Step 1:

[0525] The server collects document data, including job postings, from online platforms using APIs or scraping techniques. Inputs include URLs and access information for each platform, while output is job postings stored as text data. This data is then stored in a database for subsequent analysis.

[0526] Step 2:

[0527] The server performs vocabulary analysis on the collected text data using natural language processing libraries such as spaCy. The input is document data containing job postings, and the output is a list of words and phrases, as well as the results of contextual analysis. Here, the frequency of occurrence of specific keywords and phrases is calculated, and features that may indicate fraud are extracted.

[0528] Step 3:

[0529] The server uses a generative AI model based on TensorFlow or PyTorch to analyze fraudulent features based on the results of natural language processing. The input is the feature data obtained in step 2, and the output is a score or flag indicating the likelihood of fraudulent job postings. This score quantifies the degree of suspicion of fraud and is used in the next evaluation stage.

[0530] Step 4:

[0531] The server evaluates the trustworthiness of information providers using their past activity history and user feedback. The input is the information provider's account data and past posting history, and the output is a trustworthiness score. This score is used to determine the threshold for identifying fraudulent job postings.

[0532] Step 5:

[0533] The server notifies the user's terminal in real time of information identified as fraudulent job postings and displays a warning message. The input is the ID of the job posting identified as fraudulent in the previous stage and the user's identification information, and the output is a warning message displayed on the terminal screen. This operation allows users to recognize the risk before applying for fraudulent job postings.

[0534] Step 6:

[0535] The server removes fraudulent job postings and, if necessary, reports them to administrators or law enforcement. Input is information identified as fraudulent, and output is the removal of that information or a report of the fraudulent activity. This helps maintain a safer online environment.

[0536] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0537] This invention provides a system for more effectively detecting fraudulent job postings, combining natural language processing, poster trustworthiness evaluation, and an emotion engine that recognizes user sentiment. The operation of the system will be specifically described below as an embodiment.

[0538] First, the server collects job information from social media and job search websites. This data is dynamically retrieved using APIs and scraping techniques, and any new posts are processed immediately.

[0539] The server applies natural language processing to the collected data to extract fraudulent features. This process identifies specific keywords and contexts, revealing patterns unique to fraudulent job postings. For example, the terminal identifies posts containing risky expressions such as "mass recruitment" and "immediate start."

[0540] Furthermore, the server analyzes the poster's past activity and calculates a trust score. This score is adjusted based on user feedback and the poster's history. Posts from users with low trust scores are reviewed with particular care.

[0541] A key feature of this system is its integrated emotion engine. The emotion engine recognizes the user's emotions, and the device provides appropriate warning messages and guidance. This process customizes the warning content according to the user's emotional state, designed to reduce anxiety and apprehension. For example, if the user expresses anxiety, the device will present more detailed and reassuring guidance.

[0542] As a concrete example, if a job posting contains phrases such as "high income" or "no experience necessary," and shows signs of fraud, the server will immediately detect this information and observe the decrease in its trustworthiness. The terminal will monitor the user's reaction and display an appropriate warning through the sentiment engine. This example allows users to be appropriately vigilant against fraudulent job postings and use the information with peace of mind.

[0543] This system provides an effective means to improve the security of job postings and enhance the user experience.

[0544] The following describes the processing flow.

[0545] Step 1:

[0546] The server collects job information from social media and job sites using APIs and web scraping. This ensures that the latest information is incorporated into the system in real time.

[0547] Step 2:

[0548] The server preprocesses the collected text data, removing unnecessary information to prepare it for analysis. This allows the natural language processing engine to process the data in an optimized manner.

[0549] Step 3:

[0550] The server applies natural language processing to analyze keywords and phrases with fraudulent characteristics. This involves using pre-trained algorithms to identify common patterns in fraudulent job postings.

[0551] Step 4:

[0552] The server calculates a trust score based on the poster's activity history and user feedback. This score assesses the poster's trustworthiness and helps in detecting fraudulent activity.

[0553] Step 5:

[0554] If any fraudulent characteristics are detected, the server identifies the job posting as fraudulent and automatically notifies the administrator to request its deletion or correction.

[0555] Step 6:

[0556] The server activates an emotion engine to analyze the user's emotions. This helps the server understand the user's feelings towards fraudulent job postings and consider appropriate countermeasures.

[0557] Step 7:

[0558] The device displays warning messages and suggested actions to the user based on analysis results from the emotion engine. The content of the guidance automatically changes according to the user's emotional state.

[0559] Step 8:

[0560] Users receive the displayed warnings and take appropriate action as needed. This helps maintain a safe internet environment.

[0561] Step 9:

[0562] The server records all processing results in a database, forming a feedback loop that is used to improve future models and increase detection accuracy.

[0563] (Example 2)

[0564] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0565] In recent years, the number of job postings on the internet has increased, but some of these include malicious and fraudulent job postings, posing a risk to users who mistakenly act based on this information. Furthermore, traditional systems are insufficient in detecting fraudulent information, making it difficult for users to use job postings safely. In addition, users must judge for themselves whether a job posting is fraudulent, which often causes psychological burden. Under these circumstances, there is a need for a system that improves the security of job postings and allows users to use information with peace of mind.

[0566] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0567] In this invention, the server includes means for collecting text information including job postings, means for applying natural language processing to detect fraudulent features, means for evaluating the trustworthiness of the poster, and means for recognizing the user's emotions and providing a corresponding warning message. This makes it possible to quickly evaluate the trustworthiness of job postings and provide appropriate warnings to users. As a result, users can maintain appropriate vigilance against fraudulent job postings and use the information safely.

[0568] "Job postings" refer to information about occupations and jobs that is made public to those who wish to find employment or change jobs.

[0569] "Text information" refers to digital data composed of characters, specifically data stored in the form of text or documents.

[0570] "Natural language processing" refers to the technology that enables computers to understand and analyze human language.

[0571] "Fraudulent features" refer to suspicious or questionable patterns or keywords that are not typically found in job postings.

[0572] "Contributor reliability" refers to an indicator that quantifies the reliability of the person providing the job posting based on their past behavior and reputation.

[0573] "Means of recognizing emotions" refers to the technologies and processes used to identify and respond to human emotions.

[0574] A "warning message" refers to a notification or message displayed to alert the user.

[0575] "Means of deletion or reporting" refers to processes or functions for removing inappropriate information from a system or notifying relevant authorities of inappropriate activity.

[0576] This invention is a system for ensuring the security of job postings and is implemented using a server and terminals. First, the server collects job postings from various sources on the internet. These sources include social networking services (SNS) and job posting websites. The collected information is obtained using an API or automatically collected from web pages using scraping technology.

[0577] Next, the server applies natural language processing to the collected text information to identify fraudulent features. Natural language processing uses algorithms for specific keywords and contextual analysis. This detects patterns indicating fraudulent job postings, thereby identifying fraudulent job listings.

[0578] Furthermore, the server evaluates the trustworthiness of the job posting poster. Trustworthiness is calculated based on past activity history and user feedback. Posts with low trustworthiness receive special attention and are subject to further review.

[0579] Meanwhile, the device uses emotion recognition technology to monitor user reactions and provides appropriate warning messages to the user through an emotion engine. For example, if a user expresses anxiety about a job posting, the device displays reassuring information. This allows the user to use the information with confidence.

[0580] As a concrete example of this system, let's consider a scenario where the server detects job postings containing distinctive phrases such as "no experience necessary" and "high pay." If the reliability of these job postings is low, the terminal uses an emotion engine to issue a warning to the user. The user can then review this warning and make a safer choice.

[0581] Examples of prompt statements to input into a generative AI model are as follows:

[0582] "Please explain in detail the methods used to detect fraudulent job postings using AI. Clearly state the specific algorithms and technologies used, and also touch upon the system design that takes user sentiment into consideration."

[0583] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0584] Step 1:

[0585] The server collects job postings from internet sources. Specifically, it retrieves data using APIs or employs web scraping techniques. Inputs include specified URLs or API endpoints, and the retrieved job postings are output. This information is stored in a database and used for subsequent processing.

[0586] Step 2:

[0587] The server applies natural language processing to the collected job posting text data. In this process, the collected text data is used as input, and an algorithm is used to analyze keywords and context. As part of the data processing, specific keywords and phrases are extracted, and job postings with fraudulent characteristics are output. For example, risky terms such as "mass recruitment" and "no experience necessary" are detected.

[0588] Step 3:

[0589] The server evaluates the trustworthiness of job posting posters. Input includes the poster's past activity history and user feedback. Based on this, a trustworthiness score is calculated and the result is output. During data processing, an evaluation algorithm is applied using this historical information to adjust the score. For posters with low trustworthiness, job postings are reviewed more carefully.

[0590] Step 4:

[0591] The device recognizes the user's emotions and generates warning messages using an emotion engine. The input is user response data, which is passed through an emotion recognition algorithm. This analyzes the user's emotional state and outputs a corresponding warning message based on the results. For example, if the user shows signs of anxiety, information to reassure them will be presented.

[0592] Step 5:

[0593] The server identifies job postings with fraudulent characteristics and deletes or reports them as necessary. Here, the output data from step 2 is used as input, and a process is executed to filter out information deemed fraudulent. Finally, details of deleted data and reports are output, enabling action to be taken against fraudulent job postings.

[0594] Through these steps, the system enhances the reliability of job postings and provides an environment where users can confidently conduct their job search.

[0595] (Application Example 2)

[0596] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0597] In recent years, with the proliferation of job postings on the internet, the number of victims of false or fraudulent information has increased. Such fraudulent information can pose a significant risk to job seekers, making it necessary to provide effective means of detection and warning. However, conventional methods have the challenge of not being able to detect fraudulent information in real time and provide appropriate warnings. Furthermore, there is a lack of means to provide warnings tailored to the individual circumstances and emotions of job seekers.

[0598] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0599] In this invention, the server includes means for collecting text data containing job postings, means for applying natural language processing to the text data to detect fraudulent features, and means for evaluating the trustworthiness of the poster. This enables the detection of fraudulent job postings in real time and the provision of visual and auditory warnings to users. Furthermore, by using emotion recognition technology, it is possible to customize and provide appropriate warning content according to the user's emotions, creating an environment in which job postings can be used with peace of mind.

[0600] "Job postings" refer to information provided to job seekers online or in print media, including details such as job description, salary conditions, and work location.

[0601] "Text data" refers to a collection of information composed of characters, including textual data such as job postings.

[0602] "Natural language processing" is a technology that enables computers to understand and analyze human language, and is a method for extracting useful information from text data.

[0603] "Fraudulent characteristics" refer to the properties of information that may be false or misleading, and indicate fraudulent job postings through specific keywords or patterns.

[0604] "Means" refer to the methods or devices used to achieve a specific objective, and are elements that constitute a system.

[0605] "Poster's credibility" is an evaluation of the reliability of the person providing the job posting, and is an indicator based on past activity history and feedback.

[0606] A "warning message" is a warning statement displayed to inform a user that there is potentially malicious information, and is a notification provided through visual or auditory means.

[0607] "User" refers to an individual or legal entity that views or uses job postings through this system.

[0608] "Emotional state" refers to the psychological reactions and moods that users exhibit when they receive information.

[0609] "Optimization" is the process of adjusting or improving something to its most effective or efficient state according to specific criteria.

[0610] To implement this invention, the server first collects job information publicly available on the internet. This involves using technologies that efficiently collect the latest job information from job sites and social media using web crawlers and APIs. The collected text data is analyzed by natural language processing software to detect information with fraudulent characteristics. Here, natural language processing technologies such as Amazon Comprehend can be used.

[0611] Furthermore, the server assigns a trustworthiness rating to each poster, based on their past posting history and user ratings stored in the database. Trustworthiness analysis is efficiently performed by calculating the rating score using machine learning platforms such as Amazon SageMaker.

[0612] The primary device used by users is smart glasses, which display real-time warning messages based on collected and analyzed job information via an augmented reality display. Voice feedback using Amazon Polly is also considered, designed to appropriately capture the user's attention.

[0613] For example, when a user views a job posting containing phrases like "high pay, no experience necessary," the system instantly assesses the reliability of the information and displays a warning message on the glasses such as "This job posting carries risks." Voice guidance allows users to select job postings with confidence. In this way, the system provides an innovative service that improves the user experience through fraud detection of job postings.

[0614] Examples of prompt statements to input into a generative AI model include the following:

[0615] "Try developing an application for smart glasses that analyzes new job postings, identifies signs of fraud, and warns users. How can you ensure a safe user experience as a result?"

[0616] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0617] Step 1:

[0618] The server collects job information from the internet. During collection, it uses web crawlers and APIs to retrieve the latest job postings from job sites and social media. The input is the URL or API endpoint of each job posting, and the output is the collected raw job text data. The server then organizes this data for subsequent processing steps.

[0619] Step 2:

[0620] The server applies natural language processing to the collected job posting text data. This process uses natural language processing tools such as Amazon Comprehend to extract keywords and contexts from the text data to detect fraudulent features. The input is the raw job posting text data, and the output is the text portions identified as potentially fraudulent based on the analyzed information. The server records this as a list of fraudulent features.

[0621] Step 3:

[0622] The server evaluates the trustworthiness of job posters based on the job information it has already analyzed. The server references historical data and user feedback, and generates a trustworthiness score using tools such as Amazon SageMaker. The input consists of the analyzed job information and historical database data, and the output is a numerical trustworthiness score. This trustworthiness score will be used for future warning decisions.

[0623] Step 4:

[0624] The device provides users with visual and audible warning messages. It displays warning messages in real time via an augmented reality display and provides audio feedback using Amazon Polly. Inputs are a list of fraudulent features and confidence scores, while outputs are warning messages and audio guidance for the user. This allows the device to help users feel confident receiving information even when fraudulent activity is detected.

[0625] Step 5:

[0626] The system evaluates job postings based on information provided by the user. Users refer to warning messages from their device to avoid potentially fraudulent information. The input is warning information provided by the device, and the output is the user's safe judgment. This system allows users to use job postings with greater peace of mind.

[0627] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0628] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0629] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0630] [Fourth Embodiment]

[0631] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0632] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0633] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0634] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0635] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0636] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0637] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0638] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0639] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0640] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0641] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0642] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0643] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0644] This invention provides a system for effectively and quickly detecting fraudulent job postings and taking appropriate action based on that information. The operation of the system will be described below as an embodiment.

[0645] The server first collects text data, including job postings, from social media and job search websites. This data is collected periodically using APIs and scraping techniques, so new posts are incorporated into the system almost in real time.

[0646] Next, the server applies natural language processing techniques to the collected text data to analyze posts that may have fraudulent features. This process uses a pre-trained model to detect specific keywords, phrases, and their context. For example, the terminal will be wary of posts containing phrases commonly used in fraudulent job postings, such as "high income" and "immediate payment."

[0647] The server also evaluates the trustworthiness of the poster. This evaluation uses the poster's past activity history and user feedback. For example, a poster who has frequently made fraudulent posts in the past will be given a low trustworthiness rating. This information is used to further narrow down the possibility of fraudulent job postings.

[0648] For posts deemed to be fraudulent job postings, the server will either automatically delete them or request confirmation from the social media administrator. Furthermore, it will report them to law enforcement agencies as necessary. To prevent users from applying to fraudulent job postings, their devices will display warning messages and take appropriate measures.

[0649] For example, if a job posting containing phrases like "No experience necessary! Immediate payment, high income" is posted on a social networking site, the server will immediately detect this information and determine that it is highly likely to be fraudulent. Subsequently, the post will be promptly deleted, and the necessary information will be provided to law enforcement. During this process, a warning will be displayed on the user's device, prompting them to take action to maintain their safety.

[0650] Such systems enable the rapid detection and response to fraudulent job postings, thereby improving the security of the internet environment.

[0651] The following describes the processing flow.

[0652] Step 1:

[0653] The server collects job information as text data from social media and job search websites. This collection is performed continuously in real time using APIs and web scraping techniques.

[0654] Step 2:

[0655] The server begins preprocessing the collected text data. This process removes unnecessary HTML tags and advertisements, generating pure text data optimized for analysis.

[0656] Step 3:

[0657] The server uses a natural language processing engine to analyze text data and extract specific keywords and phrases. For example, it checks whether terms like "high income" or "same-day payment" are included.

[0658] Step 4:

[0659] The server performs contextual analysis to understand the overall tone and intent of the job posting. This involves using natural language processing algorithms to assess the potential for fraud based on the flow and structure of the sentences.

[0660] Step 5:

[0661] The server evaluates the trustworthiness of the poster. Based on past posting history and user feedback, it calculates a trustworthiness score and pays particular attention to posts with low scores.

[0662] Step 6:

[0663] The server triggers an automated deletion process based on information it identifies as fraudulent job postings. If necessary, it sends a notification to the social media administrator requesting their confirmation.

[0664] Step 7:

[0665] If the server determines that it is necessary to notify law enforcement, it will initiate an automated notification process and provide the necessary information.

[0666] Step 8:

[0667] The device displays a warning message to the user regarding fraudulent job postings. The user can review this warning and receive instructions on how to take safety precautions.

[0668] Step 9:

[0669] The server saves the results of the above processing to a database and uses them for future analysis and improvement. This forms a feedback loop to improve the accuracy of the model.

[0670] (Example 1)

[0671] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0672] The increase in fraudulent information online, particularly in job postings, is raising the risk of users becoming victims of scams and unethical recruitment practices. Such fraudulent information can lead to the leakage of users' personal information and financial losses, making it crucial to efficiently and quickly detect such information and implement appropriate countermeasures.

[0673] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0674] In this invention, the server includes means for collecting information from an information network using data communication technology, means for analyzing fraudulent features by applying natural language processing technology, and means for evaluating the trustworthiness of information providers and generating an evaluation score based on that trustworthiness. This enables the rapid detection of fraudulent information, appropriate deletion or notification, and warning display to users, thereby providing safe and reliable information.

[0675] "Data communication technology" refers to technologies that enable the transmission and reception of data over information networks.

[0676] An "information network" is a network that allows computers to exchange data with each other, including the internet and LANs.

[0677] "Natural language processing technology" is a technology that enables computers to understand and analyze human language.

[0678] "Fraudulent features" refer to specific patterns or linguistic elements that may indicate fraudulent information.

[0679] An "information provider" refers to an individual or organization that disseminates information on an information network.

[0680] A "trust score" is a quantitative measure of an information provider's reliability, calculated based on their past behavior and evaluations.

[0681] "Means for generating evaluation scores" refers to algorithms and processes for quantifying the reliability of information providers.

[0682] "Deletion or notification" refers to actions that remove malicious information from a network or alert those involved.

[0683] A "warning message" is a message displayed on a user's device that informs them that certain information may be fraudulent.

[0684] This invention is a system aimed at quickly and effectively detecting fraudulent job postings on an information network and taking appropriate measures. Specific embodiments of the system are described below.

[0685] The server first uses APIs and scraping techniques to collect text data, including job postings, from social media and job search websites. This collected data is updated in near real-time, and new information is incorporated into the system.

[0686] Next, the server applies natural language processing techniques to the collected text data. In this step, a generative AI model is used to detect keywords and phrases with specific malicious characteristics. For example, it pays attention to posts that contain phrases such as "high income" or "instant payment."

[0687] Furthermore, the server evaluates the trustworthiness of the information provider. This evaluation takes into account the information provider's past activity history and user feedback. Based on the evaluation results, a trustworthiness score is calculated for the information provider and used to detect fraudulent information.

[0688] If information is deemed fraudulent, the server will automatically delete it or notify relevant authorities. Additionally, the terminal will display a warning message to the user, prompting them to take action to ensure their security.

[0689] As a concrete example, consider a scenario where a user finds a job posting on social media with content such as "No experience necessary! Immediate payment, high income." The server immediately identifies this information as fraudulent and deletes it, and a warning message appears on the user's device stating, "This job posting may be fraudulent. Please be careful."

[0690] This invention makes it possible to quickly detect and respond to fraudulent job postings on the internet, thereby protecting the safety of users.

[0691] Examples of prompt messages are as follows:

[0692] "Please explain in detail how to detect social media posts containing fraudulent job postings in real time."

[0693] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0694] Step 1:

[0695] Data collection

[0696] The server uses APIs or scraping techniques to collect text data containing job postings from social media and job sites. The input to this process is a specified URL or hashtag from a designated information network, and the output is a list of the collected text data. For example, the server periodically calls the Twitter API to retrieve posts containing the "job postings" tag and saves them to a database.

[0697] Step 2:

[0698] Text analysis

[0699] The server applies natural language processing techniques to the collected text data. The input for this step is the text data obtained in the previous step, and the output is the analysis result indicating whether or not it is potentially fraudulent. By using a generative AI model to detect specific keywords and phrases, the characteristics of fraudulent job postings are identified. For example, posts containing expressions such as "high income" and "immediate payment" are targeted for attention.

[0700] Step 3:

[0701] Confidence Rating

[0702] The server evaluates the trustworthiness of information providers who post text data. The input for this process is data about the poster's past activity history and user feedback, and the output is a trustworthiness score. If a provider has a history of making fraudulent posts, the server assigns them a low trustworthiness score.

[0703] Step 4:

[0704] Identifying and responding to fraudulent information

[0705] The server identifies job postings deemed fraudulent and takes appropriate action. The inputs for this step are analysis results and confidence scores, while the outputs are deletion orders and notification orders. The server either automatically deletes the fraudulent information or initiates a protocol to notify the administrator. It also notifies law enforcement agencies where appropriate.

[0706] Step 5:

[0707] User notifications

[0708] The terminal displays a warning message to the user. The input to this process is a warning notification sent from the server, and the output is a warning message displayed on the screen. A message such as "This job posting may be fraudulent. Please be careful." is displayed to prevent the user from accidentally applying for fraudulent information.

[0709] (Application Example 1)

[0710] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0711] Online platforms frequently receive fraudulent job postings, leading to users mistakenly applying to these jobs. This increases the risk of users becoming victims of fraud or suffering other disadvantages. Furthermore, it is difficult to quickly identify and address fraudulent job postings, highlighting the need for effective measures to ensure user safety.

[0712] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0713] In this invention, the server includes means for collecting document data containing job postings, means for applying language processing to the document data to detect fraudulent features, and means for evaluating the trustworthiness of the information provider. This enables the rapid detection of fraudulent job postings and the issuance of warnings to the user's terminal, thereby reducing the risk of users becoming involved in fraudulent job postings and allowing them to conduct job-seeking activities safely.

[0714] "Job postings" refer to information regarding job openings and employment conditions.

[0715] "Document data" refers to all electronically stored information, including characters, symbols, and descriptions.

[0716] "Language processing" refers to the technology that enables computers to understand, analyze, and process natural language.

[0717] "Fraudulent features" refer to characteristics of information that are judged to contain intent to commit fraud or misuse.

[0718] An "information provider" refers to an entity that provides or publishes information online.

[0719] "Trustworthiness" refers to a measure used to evaluate the reliability and credibility of information providers and their information.

[0720] A "reliability score" refers to a numerical representation of trustworthiness calculated based on the information provider's past actions and feedback.

[0721] To realize this invention, a system is built through the cooperation of a server and a user's terminal. The server first collects document data, including job postings. This is typically done using APIs or scraping techniques. Next, the server applies natural language processing techniques to the collected document data to detect fraudulent features. A natural language processing library (e.g., spaCy) is used for this language processing. To identify information with fraudulent features, generative AI models using TensorFlow or PyTorch are utilized.

[0722] Furthermore, the server evaluates the trustworthiness of information providers and generates a trustworthiness score based on past activity history and user feedback. Based on this evaluation criterion, fraudulent job postings can be quickly identified. Job postings deemed fraudulent are automatically deleted or reported by the server. In addition, warnings are displayed on the user's device in real time, prompting them to take measures to prevent damage from fraudulent job postings.

[0723] For example, if a job posting such as "No experience necessary, immediate payment, high income" is displayed on a device, the server immediately analyzes the content and, if it determines that there is a possibility of fraud, displays a warning to the user saying, "Suspicious job posting has been detected. Please investigate carefully before applying."

[0724] Examples of prompts for a generative AI model:

[0725] "Please evaluate the likelihood that the following job posting is fraudulent: 'No experience necessary! Immediate payment, high income.'"

[0726] This system allows users to safely conduct job-seeking activities online.

[0727] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0728] Step 1:

[0729] The server collects document data, including job postings, from online platforms using APIs or scraping techniques. Inputs include URLs and access information for each platform, while output is job postings stored as text data. This data is then stored in a database for subsequent analysis.

[0730] Step 2:

[0731] The server performs vocabulary analysis on the collected text data using natural language processing libraries such as spaCy. The input is document data containing job postings, and the output is a list of words and phrases, as well as the results of contextual analysis. Here, the frequency of occurrence of specific keywords and phrases is calculated, and features that may indicate fraud are extracted.

[0732] Step 3:

[0733] The server uses a generative AI model based on TensorFlow or PyTorch to analyze fraudulent features based on the results of natural language processing. The input is the feature data obtained in step 2, and the output is a score or flag indicating the likelihood of fraudulent job postings. This score quantifies the degree of suspicion of fraud and is used in the next evaluation stage.

[0734] Step 4:

[0735] The server evaluates the trustworthiness of information providers using their past activity history and user feedback. The input is the information provider's account data and past posting history, and the output is a trustworthiness score. This score is used to determine the threshold for identifying fraudulent job postings.

[0736] Step 5:

[0737] The server notifies the user's terminal in real time of information identified as fraudulent job postings and displays a warning message. The input is the ID of the job posting identified as fraudulent in the previous stage and the user's identification information, and the output is a warning message displayed on the terminal screen. This operation allows users to recognize the risk before applying for fraudulent job postings.

[0738] Step 6:

[0739] The server removes fraudulent job postings and, if necessary, reports them to administrators or law enforcement. Input is information identified as fraudulent, and output is the removal of that information or a report of the fraudulent activity. This helps maintain a safer online environment.

[0740] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0741] This invention provides a system for more effectively detecting fraudulent job postings, combining natural language processing, poster trustworthiness evaluation, and an emotion engine that recognizes user sentiment. The operation of the system will be specifically described below as an embodiment.

[0742] First, the server collects job information from social media and job search websites. This data is dynamically retrieved using APIs and scraping techniques, and any new posts are processed immediately.

[0743] The server applies natural language processing to the collected data to extract fraudulent features. This process identifies specific keywords and contexts, revealing patterns unique to fraudulent job postings. For example, the terminal identifies posts containing risky expressions such as "mass recruitment" and "immediate start."

[0744] Furthermore, the server analyzes the poster's past activity and calculates a trust score. This score is adjusted based on user feedback and the poster's history. Posts from users with low trust scores are reviewed with particular care.

[0745] A key feature of this system is its integrated emotion engine. The emotion engine recognizes the user's emotions, and the device provides appropriate warning messages and guidance. This process customizes the warning content according to the user's emotional state, designed to reduce anxiety and apprehension. For example, if the user expresses anxiety, the device will present more detailed and reassuring guidance.

[0746] As a concrete example, if a job posting contains phrases such as "high income" or "no experience necessary," and shows signs of fraud, the server will immediately detect this information and observe the decrease in its trustworthiness. The terminal will monitor the user's reaction and display an appropriate warning through the sentiment engine. This example allows users to be appropriately vigilant against fraudulent job postings and use the information with peace of mind.

[0747] This system provides an effective means to improve the security of job postings and enhance the user experience.

[0748] The following describes the processing flow.

[0749] Step 1:

[0750] The server collects job information from social media and job sites using APIs and web scraping. This ensures that the latest information is incorporated into the system in real time.

[0751] Step 2:

[0752] The server preprocesses the collected text data, removing unnecessary information to prepare it for analysis. This allows the natural language processing engine to process the data in an optimized manner.

[0753] Step 3:

[0754] The server applies natural language processing to analyze keywords and phrases with fraudulent characteristics. This involves using pre-trained algorithms to identify common patterns in fraudulent job postings.

[0755] Step 4:

[0756] The server calculates a trust score based on the poster's activity history and user feedback. This score assesses the poster's trustworthiness and helps in detecting fraudulent activity.

[0757] Step 5:

[0758] If any fraudulent characteristics are detected, the server identifies the job posting as fraudulent and automatically notifies the administrator to request its deletion or correction.

[0759] Step 6:

[0760] The server activates an emotion engine to analyze the user's emotions. This helps the server understand the user's feelings towards fraudulent job postings and consider appropriate countermeasures.

[0761] Step 7:

[0762] The device displays warning messages and suggested actions to the user based on analysis results from the emotion engine. The content of the guidance automatically changes according to the user's emotional state.

[0763] Step 8:

[0764] Users receive the displayed warnings and take appropriate action as needed. This helps maintain a safe internet environment.

[0765] Step 9:

[0766] The server records all processing results in a database, forming a feedback loop that is used to improve future models and increase detection accuracy.

[0767] (Example 2)

[0768] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0769] In recent years, the number of job postings on the internet has increased, but some of these include malicious and fraudulent job postings, posing a risk to users who mistakenly act based on this information. Furthermore, traditional systems are insufficient in detecting fraudulent information, making it difficult for users to use job postings safely. In addition, users must judge for themselves whether a job posting is fraudulent, which often causes psychological burden. Under these circumstances, there is a need for a system that improves the security of job postings and allows users to use information with peace of mind.

[0770] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0771] In this invention, the server includes means for collecting text information including job postings, means for applying natural language processing to detect fraudulent features, means for evaluating the trustworthiness of the poster, and means for recognizing the user's emotions and providing a corresponding warning message. This makes it possible to quickly evaluate the trustworthiness of job postings and provide appropriate warnings to users. As a result, users can maintain appropriate vigilance against fraudulent job postings and use the information safely.

[0772] "Job postings" refer to information about occupations and jobs that is made public to those who wish to find employment or change jobs.

[0773] "Text information" refers to digital data composed of characters, specifically data stored in the form of text or documents.

[0774] "Natural language processing" refers to the technology that enables computers to understand and analyze human language.

[0775] "Fraudulent features" refer to suspicious or questionable patterns or keywords that are not typically found in job postings.

[0776] "Contributor reliability" refers to an indicator that quantifies the reliability of the person providing the job posting based on their past behavior and reputation.

[0777] "Means of recognizing emotions" refers to the technologies and processes used to identify and respond to human emotions.

[0778] A "warning message" refers to a notification or message displayed to alert the user.

[0779] "Means of deletion or reporting" refers to processes or functions for removing inappropriate information from a system or notifying relevant authorities of inappropriate activity.

[0780] This invention is a system for ensuring the security of job postings and is implemented using a server and terminals. First, the server collects job postings from various sources on the internet. These sources include social networking services (SNS) and job posting websites. The collected information is obtained using an API or automatically collected from web pages using scraping technology.

[0781] Next, the server applies natural language processing to the collected text information to identify fraudulent features. Natural language processing uses algorithms for specific keywords and contextual analysis. This detects patterns indicating fraudulent job postings, thereby identifying fraudulent job listings.

[0782] Furthermore, the server evaluates the trustworthiness of the job posting poster. Trustworthiness is calculated based on past activity history and user feedback. Posts with low trustworthiness receive special attention and are subject to further review.

[0783] Meanwhile, the device uses emotion recognition technology to monitor user reactions and provides appropriate warning messages to the user through an emotion engine. For example, if a user expresses anxiety about a job posting, the device displays reassuring information. This allows the user to use the information with confidence.

[0784] As a concrete example of this system, let's consider a scenario where the server detects job postings containing distinctive phrases such as "no experience necessary" and "high pay." If the reliability of these job postings is low, the terminal uses an emotion engine to issue a warning to the user. The user can then review this warning and make a safer choice.

[0785] Examples of prompt statements to input into a generative AI model are as follows:

[0786] "Please explain in detail the methods used to detect fraudulent job postings using AI. Clearly state the specific algorithms and technologies used, and also touch upon the system design that takes user sentiment into consideration."

[0787] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0788] Step 1:

[0789] The server collects job postings from internet sources. Specifically, it retrieves data using APIs or employs web scraping techniques. Inputs include specified URLs or API endpoints, and the retrieved job postings are output. This information is stored in a database and used for subsequent processing.

[0790] Step 2:

[0791] The server applies natural language processing to the collected job posting text data. In this process, the collected text data is used as input, and an algorithm is used to analyze keywords and context. As part of the data processing, specific keywords and phrases are extracted, and job postings with fraudulent characteristics are output. For example, risky terms such as "mass recruitment" and "no experience necessary" are detected.

[0792] Step 3:

[0793] The server evaluates the trustworthiness of job posting posters. Input includes the poster's past activity history and user feedback. Based on this, a trustworthiness score is calculated and the result is output. During data processing, an evaluation algorithm is applied using this historical information to adjust the score. For posters with low trustworthiness, job postings are reviewed more carefully.

[0794] Step 4:

[0795] The device recognizes the user's emotions and generates warning messages using an emotion engine. The input is user response data, which is passed through an emotion recognition algorithm. This analyzes the user's emotional state and outputs a corresponding warning message based on the results. For example, if the user shows signs of anxiety, information to reassure them will be presented.

[0796] Step 5:

[0797] The server identifies job postings with fraudulent characteristics and deletes or reports them as necessary. Here, the output data from step 2 is used as input, and a process is executed to filter out information deemed fraudulent. Finally, details of deleted data and reports are output, enabling action to be taken against fraudulent job postings.

[0798] Through these steps, the system enhances the reliability of job postings and provides an environment where users can confidently conduct their job search.

[0799] (Application Example 2)

[0800] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0801] In recent years, with the proliferation of job postings on the internet, the number of victims of false or fraudulent information has increased. Such fraudulent information can pose a significant risk to job seekers, making it necessary to provide effective means of detection and warning. However, conventional methods have the challenge of not being able to detect fraudulent information in real time and provide appropriate warnings. Furthermore, there is a lack of means to provide warnings tailored to the individual circumstances and emotions of job seekers.

[0802] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0803] In this invention, the server includes means for collecting text data containing job postings, means for applying natural language processing to the text data to detect fraudulent features, and means for evaluating the trustworthiness of the poster. This enables the detection of fraudulent job postings in real time and the provision of visual and auditory warnings to users. Furthermore, by using emotion recognition technology, it is possible to customize and provide appropriate warning content according to the user's emotions, creating an environment in which job postings can be used with peace of mind.

[0804] "Job postings" refer to information provided to job seekers online or in print media, including details such as job description, salary conditions, and work location.

[0805] "Text data" refers to a collection of information composed of characters, including textual data such as job postings.

[0806] "Natural language processing" is a technology that enables computers to understand and analyze human language, and is a method for extracting useful information from text data.

[0807] "Fraudulent characteristics" refer to the properties of information that may be false or misleading, and indicate fraudulent job postings through specific keywords or patterns.

[0808] "Means" refer to the methods or devices used to achieve a specific objective, and are elements that constitute a system.

[0809] "Poster's credibility" is an evaluation of the reliability of the person providing the job posting, and is an indicator based on past activity history and feedback.

[0810] A "warning message" is a warning statement displayed to inform a user that there is potentially malicious information, and is a notification provided through visual or auditory means.

[0811] "User" refers to an individual or legal entity that views or uses job postings through this system.

[0812] "Emotional state" refers to the psychological reactions and moods that users exhibit when they receive information.

[0813] "Optimization" is the process of adjusting or improving something to its most effective or efficient state according to specific criteria.

[0814] To implement this invention, the server first collects job information publicly available on the internet. This involves using technologies that efficiently collect the latest job information from job sites and social media using web crawlers and APIs. The collected text data is analyzed by natural language processing software to detect information with fraudulent characteristics. Here, natural language processing technologies such as Amazon Comprehend can be used.

[0815] Furthermore, the server assigns a trustworthiness rating to each poster, based on their past posting history and user ratings stored in the database. Trustworthiness analysis is efficiently performed by calculating the rating score using machine learning platforms such as Amazon SageMaker.

[0816] The primary device used by users is smart glasses, which display real-time warning messages based on collected and analyzed job information via an augmented reality display. Voice feedback using Amazon Polly is also considered, designed to appropriately capture the user's attention.

[0817] For example, when a user views a job posting containing phrases like "high pay, no experience necessary," the system instantly assesses the reliability of the information and displays a warning message on the glasses such as "This job posting carries risks." Voice guidance allows users to select job postings with confidence. In this way, the system provides an innovative service that improves the user experience through fraud detection of job postings.

[0818] Examples of prompt statements to input into a generative AI model include the following:

[0819] "Try developing an application for smart glasses that analyzes new job postings, identifies signs of fraud, and warns users. How can you ensure a safe user experience as a result?"

[0820] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0821] Step 1:

[0822] The server collects job information from the internet. During collection, it uses web crawlers and APIs to retrieve the latest job postings from job sites and social media. The input is the URL or API endpoint of each job posting, and the output is the collected raw job text data. The server then organizes this data for subsequent processing steps.

[0823] Step 2:

[0824] The server applies natural language processing to the collected job posting text data. This process uses natural language processing tools such as Amazon Comprehend to extract keywords and contexts from the text data to detect fraudulent features. The input is the raw job posting text data, and the output is the text portions identified as potentially fraudulent based on the analyzed information. The server records this as a list of fraudulent features.

[0825] Step 3:

[0826] The server evaluates the trustworthiness of job posters based on the job information it has already analyzed. The server references historical data and user feedback, and generates a trustworthiness score using tools such as Amazon SageMaker. The input consists of the analyzed job information and historical database data, and the output is a numerical trustworthiness score. This trustworthiness score will be used for future warning decisions.

[0827] Step 4:

[0828] The device provides users with visual and audible warning messages. It displays warning messages in real time via an augmented reality display and provides audio feedback using Amazon Polly. Inputs are a list of fraudulent features and confidence scores, while outputs are warning messages and audio guidance for the user. This allows the device to help users feel confident receiving information even when fraudulent activity is detected.

[0829] Step 5:

[0830] The system evaluates job postings based on information provided by the user. Users refer to warning messages from their device to avoid potentially fraudulent information. The input is warning information provided by the device, and the output is the user's safe judgment. This system allows users to use job postings with greater peace of mind.

[0831] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0832] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0833] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0834] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0835] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0836] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0837] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0838] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0839] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0840] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0841] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[0842] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[0843] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0844] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[0845] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0846] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0847] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0848] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0849] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0850] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0851] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted as being incorporated by reference.

[0852] The following is further disclosed regarding the embodiments described above.

[0853] (Claim 1)

[0854] A means of collecting text data including job postings,

[0855] A means for applying natural language processing to the aforementioned text data to detect fraudulent features,

[0856] A means of evaluating the trustworthiness of the poster,

[0857] Means for identifying, removing, or reporting job postings that have fraudulent characteristics,

[0858] A system that includes this.

[0859] (Claim 2)

[0860] The system according to claim 1, characterized in that the means for detecting the fraudulent features includes an algorithm for analyzing specific keywords and contexts that indicate fraudulent job postings.

[0861] (Claim 3)

[0862] The system according to claim 1, characterized in that the means for evaluating the trustworthiness of the poster generates an evaluation score based on past activity history and user feedback.

[0863] "Example 1"

[0864] (Claim 1)

[0865] A means of collecting information from an information network using data communication technology,

[0866] A means for applying natural language processing techniques to the aforementioned information and analyzing fraudulent features,

[0867] A means for evaluating the reliability of information providers and generating an evaluation score based on that reliability,

[0868] Means for identifying and deleting or notifying information that has malicious characteristics,

[0869] A means of displaying a warning on the device to encourage users to be aware of the security of their information,

[0870] A system that includes this.

[0871] (Claim 2)

[0872] The system according to claim 1, characterized in that the means for analyzing the fraudulent features includes an algorithm for analyzing specific linguistic elements and background that indicate fraudulent information.

[0873] (Claim 3)

[0874] The system according to claim 1, characterized in that the means for evaluating the trustworthiness of the information provider generates a trustworthiness score based on past behavioral history and evaluations from users.

[0875] "Application Example 1"

[0876] (Claim 1)

[0877] A device for collecting document data including job postings,

[0878] A device that applies language processing to the aforementioned document data and detects fraudulent features,

[0879] A device for evaluating the trustworthiness of information providers,

[0880] A device that identifies and deletes or reports job postings with fraudulent characteristics,

[0881] A device that sends a warning to the user's terminal regarding fraudulent job postings,

[0882] A system that includes this.

[0883] (Claim 2)

[0884] The system according to claim 1, characterized in that the device for detecting the fraudulent features includes a method for analyzing specific words and contexts that indicate fraudulent job postings.

[0885] (Claim 3)

[0886] The system according to claim 1, characterized in that the device for evaluating the trustworthiness of the information provider generates a trustworthiness score based on past activity history and user feedback.

[0887] "Example 2 of combining an emotion engine"

[0888] (Claim 1)

[0889] Means for collecting text information, including job postings,

[0890] A means for applying natural language processing to the aforementioned text information and detecting fraudulent features,

[0891] A means of evaluating the trustworthiness of the poster,

[0892] A means of recognizing the user's emotions and providing a corresponding warning message,

[0893] Means for identifying, removing, or reporting job postings that have fraudulent characteristics,

[0894] A system that includes this.

[0895] (Claim 2)

[0896] The system according to claim 1, characterized in that the means for detecting the fraudulent features includes an algorithm for analyzing specific keywords and contexts that indicate fraud.

[0897] (Claim 3)

[0898] The system according to claim 1, characterized in that the means for evaluating the trustworthiness of the poster generates an evaluation index based on past activity history and feedback from users.

[0899] "Application example 2 when combining with an emotional engine"

[0900] (Claim 1)

[0901] A means of collecting text data including job postings,

[0902] A means for applying natural language processing to the aforementioned text data to detect fraudulent features,

[0903] A means of evaluating the trustworthiness of the poster,

[0904] A means of identifying job postings with fraudulent characteristics and providing users with warning messages visually or audibly,

[0905] A means of recognizing the user's emotional state and optimizing the content of warnings,

[0906] A system that includes this.

[0907] (Claim 2)

[0908] The system according to claim 1, characterized in that the means for detecting the fraudulent features includes an algorithm for analyzing specific keywords and contexts indicating fraudulent job postings, and the warning message is provided through an augmented reality display.

[0909] (Claim 3)

[0910] The system according to claim 1, characterized in that the means for evaluating the trustworthiness of the poster generates an evaluation score based on past activity history and feedback from users, and selects and provides an appropriate warning based on the evaluation rating information. [Explanation of Symbols]

[0911] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A device for collecting document data including job postings, A device that applies language processing to the aforementioned document data and detects fraudulent features, A device for evaluating the trustworthiness of information providers, A device that identifies and deletes or reports job postings with fraudulent characteristics, A device that sends a warning to the user's terminal regarding fraudulent job postings, A system that includes this.

2. The system according to claim 1, characterized in that the device for detecting the fraudulent features includes a method for analyzing specific words and contexts that indicate fraudulent job postings.

3. The system according to claim 1, characterized in that the device for evaluating the trustworthiness of the information provider generates a trustworthiness score based on past activity history and user feedback.

Images