Personnel abnormal behavior early warning method and system for accommodation service place

Abstract

The invention discloses a personnel abnormal behavior early warning method and system for accommodation service places. The method comprises the following steps: S1, acquiring multi-source comprehensive data of a plurality of accommodation service places; s2, potential features are mined based on a TF-IDF technology; s3, taking the basic features and the potential features of each accommodation service place as input, and outputting an abnormal behavior evaluation result of each accommodation service place; s4, performing region marking according to the abnormal behavior evaluation result; and according to a marking result, determining a target risk place and a target risk person, and performing early warning. According to the technical scheme of the embodiment of the invention, possible abnormal behaviors are mined from multiple dimensions, the accommodation service place supervision system is perfected, and the abnormal situation is warned in advance.

Description

Technical Field

[0001] This invention relates to the field of abnormal behavior early warning technology, specifically to a method and system for early warning of abnormal behavior of personnel in accommodation service venues. Background Technology

[0002] In the security management of the hotel industry, the focus is often on registering hotel guests' information, implementing security systems, and preventing and combating any abnormal behavior that may occur. Due to the large flow of people and the open nature of the hotel industry, dangerous behaviors often arise, especially in serious cases that can easily violate the law. Some operators even harbor indifferent and selfish attitudes, particularly in "new business models" such as online rental hotels and e-sports hotels, where relevant laws and regulations are incomplete, industry positioning is unclear, management systems are vague, and supervision is still in an immature state.

[0003] In recent years, due to improper or lack of management of accommodation service venues, new pressures and challenges have been brought to social management work, including social security management.

[0004] Therefore, in order to improve the local security management capabilities, there is an urgent need to provide an effective measure that can provide early warning of abnormal behavior of people in accommodation service venues. Summary of the Invention

[0005] This invention provides a method and system for early warning of abnormal behavior of personnel in accommodation service venues, which can detect possible abnormal behaviors from multiple dimensions, improve the supervision system of accommodation service venues, and provide early warning of abnormal situations.

[0006] In a first aspect, the present invention provides a method for early warning of abnormal human behavior in accommodation service venues, comprising:

[0007] S1, acquire multi-source integrated data from multiple accommodation service venues, including but not limited to: basic information of accommodation service venues, basic information of guests, and publicly available third-party data;

[0008] S2, combining the basic information of the occupants and the publicly available third-party data, filter out key personnel of interest, collect the social information of the key personnel of interest, and mine the potential features in the social information based on TF-IDF technology;

[0009] S3, Select multiple basic features from the multi-source integrated data, input the basic features and the potential features together into the preset abnormal behavior risk identification model, and output the abnormal behavior assessment results of each accommodation service venue;

[0010] S4. Based on the abnormal behavior assessment results and the basic information of the accommodation service venue, identify the target risk venues and target risk personnel, and issue an early warning.

[0011] Furthermore, step S2 specifically includes:

[0012] S201, Based on the basic information of the occupants and their registered ages, a preliminary screening of potential targets is conducted; these potential targets are then grouped, and based on the grouping results, a first category of key personnel for monitoring is identified.

[0013] S202, from the publicly available third-party data, perform personal information matching on the co-residents of the target person to be monitored, and query whether there are any police-related personnel among the co-residents; based on the query results, mark the second category of key personnel to be monitored from among the co-residents;

[0014] S203, From the publicly available third-party data, text extraction is performed on the social information of the first type of key personnel and the second type of key personnel using TF-IDF technology to uncover the potential features in the social information.

[0015] Furthermore, if grouped by "having cohabitants and not having cohabitants," the specific process for marking the first category of key personnel is as follows:

[0016] S2011, determine whether the person to be monitored is a missing person, match the person to be monitored with publicly available third-party data, and if the person is a missing person, mark the person to be monitored as a first-category key person to be monitored directly;

[0017] S2012 classifies individuals under observation into two categories: those with cohabiting family members and those without. If an individual under observation has no cohabiting family members, they are marked as a non-priority individual. If an individual under observation has cohabiting family members, and those family members are relatives of the individual under observation, they are marked as a non-priority individual. If an individual under observation has cohabiting family members, and those family members are not relatives of the individual under observation, they are marked as a first-category priority individual.

[0018] Furthermore, the basic characteristics include, but are not limited to: number of beds, number of years of operation, average annual number of occupancy, number of nighttime occupancy, number of late-night occupancy, number of occupancy of the first category of key personnel and number of occupancy of the second category of key personnel;

[0019] The potential features include, but are not limited to: popular keywords in comments, sentiment trends, and topics of interest.

[0020] Furthermore, the abnormal behavior risk identification model comprehensively scores each accommodation service venue based on the pre-assigned weights of each feature in the basic features and potential features;

[0021] The abnormal behavior risk identification model pre-sets multiple candidate algorithms, calculates the performance index of each candidate algorithm using PyCaret, and selects the target algorithm from the candidate algorithms; the abnormal behavior risk identification model is trained by setting training datasets and test datasets.

[0022] The candidate algorithms include, but are not limited to: Gradient Boosting, Ada Boost, LightGBM, Logistic Regression, and Ridge.

[0023] Furthermore, step S4 specifically includes:

[0024] Based on the abnormal behavior assessment results and the basic information of the accommodation service venues, each accommodation service venue is marked as an area; based on the marking results, target risk venues are selected, and key personnel in the target risk venues are designated as target risk personnel; warnings are issued for the target risk venues and the target risk personnel.

[0025] Furthermore, the process for determining the target risk location is as follows:

[0026] Based on the assessment results of abnormal behavior of each accommodation service venue, venues are marked with different risk levels; according to the concentration of the same mark, at least one warning circle is drawn for accommodation service venues with the same mark and marked with color; the venue marks, warning circles and color marks of each warning circle are updated in real time, and the high-risk accommodation service venues corresponding to the high-risk warning circle are designated as target risk venues.

[0027] Secondly, the present invention provides an early warning system for abnormal human behavior in accommodation service venues, comprising:

[0028] The acquisition unit is used to acquire multi-source integrated data from multiple accommodation service venues. The multi-source integrated data includes, but is not limited to: basic information of the accommodation service venues, basic information of the guests, and publicly available third-party data.

[0029] The mining unit is used to combine the basic information of the residents and the publicly available third-party data to filter out key personnel of interest, collect the social information of the key personnel of interest, and mine the potential features in the social information based on TF-IDF technology.

[0030] The evaluation unit is used to select multiple basic features from the multi-source integrated data, input the basic features and the potential features into a preset abnormal behavior risk identification model, and output the abnormal behavior evaluation results of each accommodation service venue.

[0031] The early warning unit is used to identify target risk locations and target risk personnel based on the abnormal behavior assessment results and the basic information of the accommodation service venue, and to issue an early warning.

[0032] Thirdly, embodiments of the present invention provide an electronic device, the electronic device comprising:

[0033] At least one processor; and a memory communicatively connected to the at least one processor; wherein,

[0034] The memory stores a computer program that can be executed by at least one processor, such that the at least one processor is able to perform the steps of the method for early warning of abnormal human behavior in accommodation service locations according to any embodiment of the present invention.

[0035] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing computer instructions, which are used to cause a processor to execute the steps of the method for early warning of abnormal human behavior in accommodation service venues according to any embodiment of the present invention.

[0036] Compared with the prior art, the present invention has the following advantages:

[0037] The technical solution in this invention utilizes multi-dimensional data statistics and target algorithms to identify potential abnormal behavior among guests and in accommodation establishments, particularly those involving illegal activities. This further prevents abnormal events, provides early warnings of unusual situations, and more effectively promotes risk control and local management capabilities, thereby preventing abnormal events from occurring in accommodation establishments. Attached Figure Description

[0038] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only preferred embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0039] Figure 1 This is a flowchart illustrating a method for early warning of abnormal human behavior in accommodation service venues, provided in Embodiment 1 of the present invention.

[0040] Figure 2This is a schematic diagram of a system for early warning of abnormal human behavior in accommodation service venues, provided in Embodiment 2 of the present invention.

[0041] Figure 3 This is a schematic diagram of the structure of the electronic device provided in Embodiment 3 of the present invention. Detailed Implementation

[0042] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0043] Example 1

[0044] Figure 1 This is a flowchart illustrating a method for early warning of abnormal human behavior in accommodation service venues, as provided in Embodiment 1 of the present invention. This embodiment is applicable to situations where risk warnings are issued for accommodation service venues and their occupants. The method can be executed by an early warning system for abnormal human behavior in accommodation service venues, which can be implemented using software and / or hardware.

[0045] like Figure 1 As shown, the method specifically includes:

[0046] S1, acquire multi-source integrated data from multiple accommodation service venues. The multi-source integrated data includes, but is not limited to: basic information of accommodation service venues, basic information of guests, and publicly available third-party data.

[0047] The basic information of accommodation service venues includes: name and logo, geographical location, contact information, business hours, room type and occupancy, public facilities, guest room facilities, services, price information, safety measures and check-in instructions; the basic information of guests includes: name, gender, age, de-identified travel records, contact information; and publicly available third-party data includes: publicly available social information of individuals, police information and missing persons information disclosed by third-party platforms.

[0048] It should be noted that during the data collection process, due to the massive amount of data, it is necessary to extract representative feature data from the multi-source integrated data, and these feature data need to be numerically processed.

[0049] One of the core aspects of this invention is: selecting representative basic features from multi-source integrated data, and mining representative potential features from multi-source integrated data, quantifying these basic and potential features, and inputting them together into a pre-built abnormal behavior risk identification model to identify occupants and accommodation service locations that may exhibit abnormal behavior.

[0050] Optionally, the multi-source integrated data can be preprocessed to summarize statistical information such as cohabitants, cohabitation frequency, and number of cohabitants, extract statistical information from multiple dimensions, map the statistical data to the range [0, 1], and mark them as different weight values ​​to avoid errors caused by different units.

[0051] To achieve the acquisition of comprehensive data from multiple sources, an internal system platform was built, and automated tools and interfaces were developed to acquire relevant data information from different sources in real time, and to integrate and clean the data to ensure data consistency and prepare for the standardization of feature data in subsequent algorithms.

[0052] S2 combines the basic information of the residents with publicly available third-party data to identify key individuals for monitoring, and collects their social information, using TF-IDF technology to uncover potential features within that social information.

[0053] Furthermore, step S2 specifically includes:

[0054] S201. Based on the basic information of the occupants and their registered ages, a preliminary screening of individuals to be monitored is conducted. These individuals are then grouped, and based on the grouping results, the first category of individuals requiring special attention is identified.

[0055] In this embodiment of the invention, minors are considered as one of the target individuals to be monitored. The target individuals to be monitored can be initially screened based on the age registered by the residents. They can be grouped according to categories such as "local and out-of-town", "with and without cohabiting persons", and "cohabiting with opposite sexes and cohabiting with same sexes".

[0056] If grouped by "having cohabitants and not having cohabitants", the specific process for marking the first category of key personnel is as follows:

[0057] S2011, determine whether the person to be monitored is a missing person, match the person to be monitored with publicly available third-party data, and if the person is a missing person, mark the person to be monitored as a first-category key person to be monitored.

[0058] S2012 classifies individuals under observation into two categories: those with cohabiting family members and those without. If an individual under observation has no cohabiting family members, they are marked as a non-priority individual. If an individual under observation has cohabiting family members, and those family members are relatives of the individual under observation, they are marked as a non-priority individual. If an individual under observation has cohabiting family members, and those family members are not relatives of the individual under observation, they are marked as a first-category priority individual.

[0059] Among them, co-residents refer to persons who register and check in together with the person under observation. The kinship between co-residents and the person under observation can be: the co-resident is the guardian or other relative of the person under observation.

[0060] Similarly, the "local" dataset can be matched with resident population information to identify individuals not registered in the same household, further filtering out those whose kinship cannot be determined. After re-verification, suspected erroneous data is reported and their accommodation status is investigated. The "other city" dataset is searched in the national population information system and compared with the guardian information registered by people from other cities. Individuals with mismatches are marked as anomalies. Individuals with erroneous registration information are designated as the first category of key individuals requiring attention.

[0061] Similarly, the "cohabitation with opposite sex" dataset was filtered using a data collision method to separate data into "no kinship" and "kinship" categories, providing feedback on cohabitation situations where individuals are not related by blood. Based on the registration data, individuals recorded under categories such as "couples" or "lovers" were highlighted and reported back. These individuals cohabiting with opposite sexes who are not related by blood were designated as the first category of key individuals requiring attention.

[0062] S202: Based on publicly available third-party data, personal information of cohabitants of the person under investigation is matched to check whether any of them are involved with the police; based on the results, the second category of key individuals under investigation is identified from among the cohabitants.

[0063] Among them, "personnel involved in police affairs" refers to those with prior criminal records or administrative penalty records.

[0064] After initially identifying the individuals to be monitored, the first category of key individuals to be monitored is marked, and then the second category of key individuals to be monitored are marked among the cohabitants of the individuals to be monitored.

[0065] S203. From publicly available third-party data, text extraction is performed on the social information of the first and second categories of key personnel using TF-IDF technology to uncover potential features in the social information.

[0066] Potential features include: trending keywords in comments, sentiment trends, and areas of interest. By mining comments and trending topics of interest from social media, combined with basic information about residents, and increasing statistics on interaction frequency and number of participants, the system extracts key information from large amounts of text using TF-IDF. Analyzing the trending keywords, sentiment trends, and areas of interest of key individuals allows for an understanding of residents' behavioral tendencies. In particular, words, homophones, and antonyms with potentially harmful connotations are given higher weights after TF-IDF text processing.

[0067] TF-IDF is a core feature extraction technique in text processing. By quantifying the importance of words in a document, it transforms unstructured text into a structured numerical matrix, converting text into numerical features. Hot words and trending topics in comments are extracted from the comment text, quantified into numerical features, and assigned higher weights to harmful or negative information. Sentiment analysis is based on a sentiment lexicon. After segmenting the comment text, TF-IDF is used to calculate and statistically analyze sentiment words, assigning 1 to positive words, 0 to neutral words, and -1 to negative words, thus calculating sentiment scores for different individuals.

[0068] S3 selects multiple basic features from multi-source integrated data, inputs the basic features and potential features into the preset abnormal behavior risk identification model, and outputs the abnormal behavior assessment results of each accommodation service venue.

[0069] It should be noted that after obtaining the basic features and potential features, these features need to be numerically processed first. The model inputs in this embodiment are all numerical data.

[0070] In this embodiment of the invention, seven basic features were selected from multi-source integrated data, including: number of beds, number of years of operation, average annual number of occupancy, number of occupancy at night (22:00 to 24:00), number of occupancy at midnight (0:00 to 4:00), number of occupancy of the first category of key personnel, and number of occupancy of the second category of key personnel; three potential features were mined from multi-source integrated data, including: popular comment terms, sentiment tendencies, and hot topics of interest.

[0071] Optionally, different weights can be assigned to each feature in the basic and potential features based on its correlation with abnormal behavior. Among them, the average annual number of guests, the number of nighttime guests, and the number of late-night guests are strong correlation factors for abnormal behavior in traditional accommodation service venues and can be assigned higher weights; while the number of years of operation, the number of beds, the number of guests of the first category of key attention, the number of guests of the second category of key attention, popular comments, sentiment tendencies, and hot topics of concern are all weak correlation factors and can be assigned lower weights.

[0072] Optionally, the abnormal behavior risk identification model can comprehensively score each accommodation service venue based on the pre-assigned weights of each feature in the basic and potential features.

[0073] Optionally, multiple candidate algorithms are pre-set in the abnormal behavior risk identification model. The performance index of each candidate algorithm is calculated using PyCaret, and the target algorithm is selected from the candidate algorithms. The abnormal behavior risk identification model is trained by setting up training datasets and test datasets.

[0074] Different algorithms need to be selected depending on the basic and potential features to ensure the accuracy of abnormal behavior assessment results.

[0075] PyCaret is an open-source, low-code Python machine learning library designed to simplify and accelerate machine learning workflows. It integrates various commonly used machine learning algorithms and tools, automating tasks such as data preprocessing, model training, tuning, evaluation, and deployment. PyCaret supports a wide range of machine learning tasks, including classification, regression, clustering, anomaly detection, natural language processing (NLP), and association rule mining. PyCaret provides a built-in feature importance assessment tool to help understand which features are most important.

[0076] Commonly used classification algorithms for comparison include: Gradient Boosting, Ada Boost, LightGBM, Logistic Regression, and Ridge. Gradient Boosting optimizes the loss function through gradient descent and iteratively fits the residuals; its main feature is the ability to define a custom loss function. Ada Boost iteratively adjusts sample weights and sequentially trains weak classifiers (such as decision stumps). LightGBM is an efficient gradient boosting framework primarily used for classification, regression, and ranking problems, especially excelling in handling large-scale data. LightGBM boasts high training speed, supports large-scale data, and its accuracy is comparable to or better than other gradient boosting frameworks (such as XGBoost), particularly in multi-class classification and regression tasks. Logistic Regression is a linear model that maps linear outputs to probabilities using the sigmoid function. Ridge is an L2-regularized linear regression, similar to Logistic Regression + L2 regularization when used for classification.

[0077] In this embodiment of the invention, PyCaret is mainly used to evaluate the applicability of algorithms. Specifically, PyCaret code is written in a Python environment to compare the performance metrics of different algorithms. After selecting 10 features provided in this embodiment of the invention, performance metrics are calculated for 5 preset algorithms based on these 10 features. Table 1 is a ranking table of candidate algorithms and their corresponding performance metrics, including: accuracy, ROC curve, recall, precision, F1 score, Kappa, Matthews correlation coefficient, etc. Various performance metrics of the model are comprehensively evaluated, and a suitable algorithm is selected as the target algorithm from the candidate algorithms. The compare_models function returns a table containing performance comparisons of all algorithms, sorted by default metrics (such as accuracy). For example, the following five algorithms are selected for prediction comparison: Gradient Boosting, AdaBoost, LightGBM, Logistic Regression, and Ridge.

[0078] Table 1: Ranking of candidate algorithms and their corresponding performance metrics

[0079] Model Accuracy AUC Recall Precision F1 Kappa MCC Light GBM 0.74 0.81 0.62 0.75 0.67 0.47 0.47 Gradient Boosting 0.72 0.77 0.55 0.72 0.62 0.41 0.41 Logistic Regression 0.72 0.77 0.63 0.67 0.65 0.40 0.40 Ridge 0.70 0.76 0.62 0.67 0.63 0.39 0.39 Ada Boost 0.70 0.75 0.61 0.67 0.63 0.38 0.39

[0080] For five different types of algorithms, the performance metrics are as follows: Accuracy: Represents the proportion of correctly predicted cases. AUC (Area Under the ROC Curve): Used to evaluate the performance of a classification model; a value closer to 1 is better. Recall: Represents the proportion of correctly identified positive cases. Precision: Represents the proportion of samples predicted as positive that are actually positive. F1 Score: The harmonic mean of precision and recall, used to balance the two. Kappa coefficient: Used to evaluate the accuracy of a classification model, considering the impact of random guessing. MCC (Matthews correlation coefficient): Used to evaluate the performance of a binary classification model, ranging from -1 to 1. TT Sec (Training time in seconds): Represents the time required to train the model.

[0081] Based on a comparison of the performance of the different algorithms mentioned above, LightGBM (Lightweight Gradient Boosting Machine) achieved the highest accuracy. Other relevant model metrics were within acceptable ranges. Therefore, considering all factors, LightGBM is the optimal algorithm and is the target algorithm for this model. The LightGBM algorithm will be used to analyze and identify accommodation service locations that may exhibit abnormal behavior that has not yet been detected, thus identifying abnormal behaviors and potential risks.

[0082] In this embodiment of the invention, PyCaret is used to predict discrete and disordered category labels, and the applicability of the algorithms is compared. The target algorithm is used to analyze and mine potential guests and accommodation service locations that may have abnormal behavior, so as to assist in public security management and make every effort to prevent illegal events in accommodation locations.

[0083] In this embodiment of the invention, after acquiring a large number of basic and potential features, 80% is used as the training dataset and 20% as the test dataset. Data preprocessing and feature engineering are performed sequentially. The training dataset is used to train the model, resulting in a trained abnormal behavior risk identification model. Real-time data sources are used as prediction data, and the trained model classifies the prediction results.

[0084] S4. Based on the abnormal behavior assessment results and the basic information of the accommodation service venues, identify the target risk venues and target risk personnel, and issue early warnings.

[0085] Specifically, based on the results of abnormal behavior assessments and basic information about accommodation service venues, each accommodation service venue is marked as an area; based on the marking results, target risk venues are selected, and key personnel in the target risk venues are designated as target risk personnel; and warnings are issued for the target risk venues and target risk personnel.

[0086] The process for identifying target risk locations is as follows:

[0087] Based on the assessment results of abnormal behavior of each accommodation service venue, venues are marked with different risk levels; according to the concentration of the same mark, at least one warning circle is drawn for accommodation service venues with the same mark and marked with color; the venue marks, warning circles and color marks of each warning circle are updated in real time, and the high-risk accommodation service venues corresponding to the high-risk warning circle are designated as target risk venues.

[0088] In this embodiment of the invention, the abnormal behavior risk identification model outputs the abnormal behavior assessment results of each accommodation service venue. These venues can be categorized into different risk levels, such as very high, high, medium, and low risk. The concentration of different risk levels can be further classified into red, orange, yellow, and green warning zones. For example, if multiple high-risk venues exist within a certain area, that area is designated as an orange warning zone. This dynamic updating of the classification labels helps in the timely identification of abnormal behaviors and risks.

[0089] The technical solution in this embodiment of the invention is based on an abnormal behavior risk identification model to identify target risk locations and target risk personnel that may have abnormal behavior; it selects the target algorithm corresponding to the optimal performance index based on PyCaret to improve the accuracy of model identification; at the input end of the model, it selects multi-dimensional basic features and mines a variety of potential features based on TF-IDF technology to realize the mining of possible abnormal behavior from multiple dimensions, improve the supervision system of accommodation service venues, and provide early warning of abnormal situations.

[0090] Example 2

[0091] Figure 2 This is a schematic diagram of an abnormal behavior early warning system for accommodation service venues provided in Embodiment 2 of the present invention, as shown below. Figure 2 As shown, the system specifically includes:

[0092] The acquisition unit 100 is used to acquire multi-source integrated data from multiple accommodation service venues. The multi-source integrated data includes, but is not limited to: basic information of accommodation service venues, basic information of guests, and publicly available third-party data.

[0093] Mining unit 200 is used to combine the basic information of residents and publicly available third-party data to screen out key personnel of interest and collect their social information, and to mine the potential features in the social information based on TF-IDF technology.

[0094] The assessment unit 300 is used to select multiple basic features from multi-source integrated data, input the basic features and potential features into the preset abnormal behavior risk identification model, and output the abnormal behavior assessment results of each accommodation service venue.

[0095] The early warning unit 400 is used to identify target risk locations and target risk personnel based on the results of abnormal behavior assessment and basic information of accommodation service venues, and to issue early warnings.

[0096] The technical solutions in this invention identify potential risk locations and individuals with abnormal behavior through various basic features and potential signs, thereby enabling the discovery of potential abnormal behavior from multiple dimensions, improving the supervision system for accommodation service venues, and providing early warnings of abnormal situations.

[0097] Example 3

[0098] Figure 3This is a schematic diagram of an electronic device implementing the method for early warning of abnormal human behavior in accommodation service venues according to embodiments of the present invention. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (such as helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0099] like Figure 3 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 may also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0100] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0101] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, central processing unit (CPU), graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as a method for early warning of abnormal human behavior in accommodation service establishments.

[0102] In some embodiments, the method for early warning of abnormal personnel behavior in accommodation establishments can be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the method for early warning of abnormal personnel behavior in accommodation establishments described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to perform the method for early warning of abnormal personnel behavior in accommodation establishments by any other suitable means (e.g., by means of firmware).

[0103] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0104] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0105] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0106] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0107] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0108] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0109] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0110] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for early warning of abnormal personnel behavior in accommodation service establishments, characterized in that, include: S1, acquire multi-source integrated data from multiple accommodation service venues, including but not limited to: basic information of accommodation service venues, basic information of guests, and publicly available third-party data; S2, combining the basic information of the occupants and the publicly available third-party data, filter out key personnel of interest, collect the social information of the key personnel of interest, and mine the potential features in the social information based on TF-IDF technology; S3, Select multiple basic features from the multi-source integrated data, input the basic features and the potential features together into the preset abnormal behavior risk identification model, and output the abnormal behavior assessment results of each accommodation service venue; S4. Based on the abnormal behavior assessment results and the basic information of the accommodation service venue, identify the target risk venues and target risk personnel, and issue an early warning.

2. The method according to claim 1, characterized in that, Step S2 specifically includes: S201, Based on the basic information of the occupants and their registered ages, a preliminary screening of potential targets is conducted; these potential targets are then grouped, and based on the grouping results, a first category of key personnel for monitoring is identified. S202, from the publicly available third-party data, perform personal information matching on the co-residents of the target person to be monitored, and query whether there are any police-related personnel among the co-residents; based on the query results, mark the second category of key personnel to be monitored from among the co-residents; S203, From the publicly available third-party data, text extraction is performed on the social information of the first type of key personnel and the second type of key personnel using TF-IDF technology to uncover the potential features in the social information.

3. The method according to claim 2, characterized in that, If grouped by "having cohabitants and not having cohabitants", the specific process for marking the first category of key personnel is as follows: S2011, determine whether the person to be monitored is a missing person, match the person to be monitored with publicly available third-party data, and if the person is a missing person, mark the person to be monitored as a first-category key person to be monitored directly; S2012 classifies individuals under observation into two categories: those with cohabiting family members and those without. If an individual under observation has no cohabiting family members, they are marked as a non-priority individual. If an individual under observation has cohabiting family members, and those family members are relatives of the individual under observation, they are marked as a non-priority individual. If an individual under observation has cohabiting family members, and those family members are not relatives of the individual under observation, they are marked as a first-category priority individual.

4. The method according to claim 1, characterized in that, The basic characteristics include, but are not limited to: number of beds, number of years of operation, average annual number of occupancy, number of nighttime occupancy, number of late-night occupancy, number of occupancy of the first category of key personnel and number of occupancy of the second category of key personnel; The potential features include, but are not limited to: popular keywords in comments, sentiment trends, and topics of interest.

5. The method according to claim 1, characterized in that, The abnormal behavior risk identification model comprehensively scores each accommodation service venue based on the pre-assigned weights of each feature in the basic features and potential features. The abnormal behavior risk identification model pre-sets multiple candidate algorithms, calculates the performance index of each candidate algorithm using PyCaret, and selects the target algorithm from the candidate algorithms; the abnormal behavior risk identification model is trained by setting training datasets and test datasets. The candidate algorithms include, but are not limited to: Gradient Boosting, Ada Boost, LightGBM, LogisticRegression, and Ridge.

6. The method according to claim 1, characterized in that, Step S4 specifically includes: Based on the abnormal behavior assessment results and the basic information of the accommodation service venues, each accommodation service venue is marked as an area; based on the marking results, target risk venues are selected, and key personnel in the target risk venues are designated as target risk personnel; warnings are issued for the target risk venues and the target risk personnel.

7. The method according to claim 6, characterized in that, The process of determining the target risk location is as follows: Based on the assessment results of abnormal behavior at each accommodation service venue, venues are marked with different risk levels; according to the concentration of the same mark, at least one warning circle is drawn for accommodation service venues with the same mark and marked with color. The location markers, warning zones, and color markings of each accommodation service venue are updated in real time, and the high-risk accommodation service venues corresponding to the high-risk warning zones are designated as target risk venues.

8. A system for early warning of abnormal personnel behavior in accommodation service establishments, characterized in that, The system is configured to implement the method according to any one of claims 1-7, the system comprising: The acquisition unit is used to acquire multi-source integrated data from multiple accommodation service venues. The multi-source integrated data includes, but is not limited to: basic information of the accommodation service venues, basic information of the guests, and publicly available third-party data. The mining unit is used to combine the basic information of the residents and the publicly available third-party data to filter out key personnel of interest, collect the social information of the key personnel of interest, and mine the potential features in the social information based on TF-IDF technology. The evaluation unit is used to select multiple basic features from the multi-source integrated data, input the basic features and the potential features into a preset abnormal behavior risk identification model, and output the abnormal behavior evaluation results of each accommodation service venue. The early warning unit is used to identify target risk locations and target risk personnel based on the abnormal behavior assessment results and the basic information of the accommodation service venue, and to issue an early warning.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; and a memory communicatively connected to said at least one processor; wherein, The memory stores a computer program executable by the at least one processor, which enables the at least one processor to perform the steps of the method for early warning of abnormal human behavior in accommodation service locations as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that, when executed by a processor, implement the steps of the method for early warning of abnormal human behavior in accommodation service locations as described in any one of claims 1-7.

Images