Apparatus and method for detecting morphological incompleteness of self-introduction answer

Abstract

An apparatus for detecting formal incompleteness of a self-introduction response according to the present disclosure comprises: an input / output module that receives user input or provides output to the user based on a user interface; a memory in which at least one process for performing an operation to detect an error category of a sentence within the self-introduction is stored; and at least one processor that performs an operation to detect an error category of a sentence within the self-introduction according to the process. The at least one processor is configured to acquire a response sentence from the input self-introduction, detect an error category corresponding to each of the response sentences among predefined error categories, list at least one error category detected in each response sentence, and provide each response sentence and the listed error category to the input / output module. The error category may include at least one of a non-answer that is not in the form of an answer, a non-sentence that is not in the form of a sentence, a non-word that is not a word, a minor error containing punctuation that does not conform to rules, and a bullet that is a sentence written in a bulleted descriptive form.

Description

Technology Field

[0001] The present disclosure relates to a natural language processing device, and more specifically, to a device and method for detecting morphological incompleteness of a self-introduction response through natural language processing. Background Technology

[0002] Conventional recruitment systems faced the problem of incurring significant time and costs because HR personnel had to manually read applicants' documents, extract necessary information, and organize it. This was particularly problematic in the case of open recruitment by large corporations attracting over 10,000 applicants, where the human resource costs and workload for processing were substantial, and maintaining the consistency of extracted information was difficult depending on the proficiency of the HR personnel.

[0003] Therefore, answer items within the above recruitment documents must be extracted using natural language processing and preprocessed into a document of a certain format so that HR personnel can verify only the meaningful answer items.

[0004] However, if applicants do not write answers composed of proper sentences, there is a problem in that it becomes difficult to extract accurate and meaningful answer items.

[0005] Therefore, there was a need to verify whether the applicant's response items in the recruitment documents were written in the proper format, and if errors existed, to detect what kind of errors they were in order to determine the direction for future sentence correction or preprocessing.

[0006] In conventional patent literature, it is possible to provide feedback on the answer level by determining whether the interviewer satisfies the level of answer desired by the interviewer, but it is not possible to check for errors in each sentence or the error category. Prior art literature

[0007] Republic of Korea Registered Patent Publication 10-2584078 (2023.10.04) The problem to be solved

[0008] The purpose of the embodiments disclosed in this disclosure is to provide an apparatus and a method for detecting the morphological incompleteness of a response in a self-introduction letter so as to classify which of the predefined error categories a response sentence in the self-introduction letter belongs to.

[0009] The problems that this disclosure aims to solve are not limited to those mentioned above, and other unmentioned problems will be clearly understood by a person skilled in the art from the description below. means of solving the problem

[0010] An apparatus for detecting the formal incompleteness of a self-introduction response according to the present disclosure for achieving the aforementioned technical problem comprises: an input / output module that receives user input or provides output to the user based on a user interface; a memory in which at least one process for performing an operation to detect an error category of a sentence within the self-introduction is stored; and at least one processor that performs an operation to detect an error category of a sentence within the self-introduction according to the process. The at least one processor is configured to acquire a response sentence from the input self-introduction, detect an error category corresponding to each of the response sentences among predefined error categories, list at least one error category detected in each response sentence, and provide each response sentence and the listed error category to the input / output module. The error category may include at least one of a non-answer that is not in the form of an answer, a non-sentence that is not in the form of a sentence, a non-word that is not a word, a minor error containing punctuation that does not conform to rules, and a bullet that is a sentence written in a bulleted descriptive form.

[0011] Additionally, according to one embodiment of the present disclosure, the at least one processor may be configured to obtain the answer sentence by deleting a pre-set question sentence within the self-introduction and extracting the remaining sentence.

[0012] Meanwhile, according to one embodiment of the present disclosure, the at least one processor may be configured to divide the answer sentence into phrases based on spaces, and if the last word of the answer sentence is a term stored in a dictionary, or if the last word of the answer sentence is a first term stored in a dictionary and a second term stored in a dictionary is placed consecutively with the first term or separated by one word, the answer sentence may be detected as a non-answer among the error categories.

[0013] According to one embodiment of the present disclosure, the at least one processor may be configured to detect the answer sentence as a non-word among the error categories if the answer sentence includes a string of characters of a length greater than or equal to a preset length without spaces, or if the answer sentence includes a string of characters consisting of only consonants or vowels.

[0014] According to one embodiment of the present disclosure, the at least one processor may be configured to exclude the answer sentence from the non-word category if the answer sentence contains a string of consecutive characters of a length greater than or equal to a preset length without spaces, or if the string has a pattern in which a specific number of numbers and periods are repeated, or if the string consists of a specific English string followed by an English letter, a number, a period, or a diagonal line.

[0015] According to one embodiment of the present disclosure, the at least one processor may be configured to detect the answer sentence as a non-sentence among the error categories if the last word of the answer sentence is a word that is not a predicate and the last word of the answer sentence is not a form in which a common noun of a predefined modal descriptive form exists alone.

[0016] According to one embodiment of the present disclosure, the at least one processor may be further configured to process the answer sentence in natural language and classify it by morpheme, and if a morpheme other than the last morpheme of the answer sentence corresponds to a pre-final ending, and the morpheme following the pre-final ending does not correspond to at least one of a terminal ending, a linking ending, a derivational ending, a pre-final ending, a nominal derivational ending, an adjective derivational ending, an adjective derivational suffix, and an accusative case particle, the answer sentence may be further configured to detect that the answer sentence corresponds to the non-sentence category.

[0017] According to one embodiment of the present disclosure, the at least one processor may be configured to process the answer sentence in natural language and separate it by morpheme, and to detect that the answer sentence corresponds to a minor error among the error categories if the last morpheme of the answer sentence is a word that is a predicate, or if the last morpheme of the answer sentence is not another symbol corresponding to at least one of a period, an exclamation mark, and a question mark.

[0018] According to one embodiment of the present disclosure, the at least one processor may be configured to output that the answer sentence corresponds to a bullet among the error categories if the last word of the answer sentence is a word in which a common noun of a predefined modal descriptive form exists alone, and to provide the answer sentence to the input / output module to receive input regarding whether the answer sentence is a bullet or a non-sentence.

[0019] In addition, according to one embodiment of the present disclosure, the at least one processor may be configured to output only the non-sentence category if the entire sentence of the answer sentence corresponds to both the non-sentence category and the non-word category, or to output the error category corresponding to the longest sentence among the lengths of sentences corresponding to each error category if the answer sentence corresponds to a plurality of error categories.

[0020] In addition, a method for detecting formal incompleteness of a self-introduction response according to the present disclosure for achieving the aforementioned technical problem comprises: a step of obtaining a response sentence from an input self-introduction; a step of detecting an error category corresponding to each of the response sentences among predefined error categories; a step of listing at least one error category detected in each response sentence; and a step of providing each response sentence and the listed error category to an input / output module of an electronic device that receives user input or provides output to the user based on a user interface, wherein the error category may include at least one of a non-response that is not in the form of a response, a non-sentence that is not in the form of a sentence structure, a non-word that is not a word, a minor error containing punctuation that does not conform to rules, and a bullet that is a sentence written in a bulleted descriptive form.

[0021] In addition to this, a computer program stored on a computer-readable recording medium for implementing the present disclosure may be further provided.

[0022] In addition to this, a computer-readable recording medium for recording a computer program for implementing the present disclosure may be further provided. Effects of the invention

[0023] According to the aforementioned means for solving the problem of the present disclosure, an error category corresponding to an answer sentence within a self-introduction letter is detected, and the answer sentence is preprocessed according to the error category to provide the effect of improving the accuracy of answer sentence extraction.

[0024] The effects of the present disclosure are not limited to those mentioned above, and other unmentioned effects will be clearly understood by a person skilled in the art from the description below. Brief explanation of the drawing

[0025] FIG. 1 is a block diagram briefly illustrating the configuration of a device for detecting the morphological incompleteness of a self-introduction letter response according to one embodiment of the present disclosure. FIG. 2 is a block diagram briefly illustrating the process of a device for detecting the morphological incompleteness of a self-introduction letter response according to one embodiment of the present disclosure. FIG. 3 is an algorithm illustrating a Request process performed by a device for detecting the formal incompleteness of a self-introduction letter response according to one embodiment of the present disclosure. FIG. 4 is an algorithm illustrating a Response process performed by a device for detecting the formal incompleteness of a self-introduction letter response according to one embodiment of the present disclosure. FIG. 5 is a flowchart illustrating the non-answer category output process of a device for detecting the morphological incompleteness of a self-introduction letter response according to one embodiment of the present disclosure. FIG. 6 is a flowchart illustrating a non-word category output process of a device for detecting the morphological incompleteness of a self-introduction letter answer according to one embodiment of the present disclosure. FIG. 7 is a flowchart illustrating the non-sentence category output process of a device for detecting the morphological incompleteness of a self-introduction letter answer according to one embodiment of the present disclosure. FIG. 8 is a flowchart illustrating a method for detecting morphological incompleteness of a self-introduction letter response according to one embodiment of the present disclosure. Specific details for implementing the invention

[0026] Throughout this disclosure, the same reference numerals denote the same components. This disclosure does not describe all elements of the embodiments, and general content in the art to which this disclosure pertains or content that overlaps between embodiments is omitted. The terms 'part, module, component, block' as used in the specification may be implemented in software or hardware, and depending on the embodiments, a plurality of 'parts, modules, components, blocks' may be implemented as a single component, or a single 'part, module, component, block' may include a plurality of components.

[0027] Throughout the specification, when a part is described as being "connected" to another part, this includes not only cases where they are directly connected but also cases where they are indirectly connected, and indirect connections include connections made via a wireless communication network.

[0028] Furthermore, when it is stated that a part "includes" a certain component, this means that, unless specifically stated otherwise, it does not exclude other components but may include additional components.

[0029] Throughout the specification, when it is stated that a component is located "on" another component, this includes not only cases where a component is in contact with another component, but also cases where another component exists between the two components.

[0030] The terms first, second, etc. are used to distinguish one component from another, and the components are not limited by the aforementioned terms.

[0031] Singular expressions include plural expressions unless there is an obvious exception in the context.

[0032] In each step, identification codes are used for convenience of explanation and do not describe the order of the steps; the steps may be performed differently from the specified order unless a specific order is clearly indicated in the context.

[0033] The operating principles and embodiments of the present disclosure will be described below with reference to the attached drawings.

[0034] In this specification, the term "device according to the present disclosure" includes all various devices capable of performing computational processing and providing results to a user. For example, the device according to the present disclosure may include all of a computer, a server device, and a portable terminal, or may be in the form of any one of these.

[0035] Here, the computer may include, for example, a notebook, desktop, laptop, tablet PC, slate PC, etc. equipped with a web browser.

[0036] The above server device is a server that processes information by communicating with an external device, and may include an application server, a computing server, a database server, a file server, a game server, a mail server, a proxy server, and a web server.

[0037] The above portable terminal may include, for example, all types of handheld-based wireless communication devices such as PCS (Personal Communication System), GSM (Global System for Mobile communications), PDC (Personal Digital Cellular), PHS (Personal Handyphone System), PDA (Personal Digital Assistant), IMT (International Mobile Telecommunication)-2000, CDMA (Code Division Multiple Access)-2000, W-CDMA (W-Code Division Multiple Access), WiBro (Wireless Broadband Internet) terminals, smartphones, etc., as well as wearable devices such as watches, rings, bracelets, anklets, necklaces, glasses, contact lenses, or head-mounted devices (HMDs).

[0038] Functions related to artificial intelligence according to the present disclosure are operated through a processor and memory. The processor may be composed of one or more processors. In this case, the one or more processors may be general-purpose processors such as CPUs, APs, and DSPs (Digital Signal Processors), graphics-dedicated processors such as GPUs and VPUs (Vision Processing Units), or artificial intelligence-dedicated processors such as NPUs. The one or more processors control the processing of input data according to predefined operation rules or artificial intelligence models stored in memory. Alternatively, if the one or more processors are artificial intelligence-dedicated processors, the artificial intelligence-dedicated processors may be designed with a hardware structure specialized for processing a specific artificial intelligence model.

[0039] The predefined operating rules or artificial intelligence models are characterized by being created through learning. Here, being created through learning means that a predefined operating rules or artificial intelligence models configured to perform desired characteristics (or objectives) are created by a basic artificial intelligence model being trained using multiple learning data by a learning algorithm. Such learning may be performed on the device itself where the artificial intelligence according to the present disclosure is executed, or it may be performed through a separate server and / or system. Examples of learning algorithms include supervised learning, unsupervised learning, semi-supervised learning, or reinforcement learning, but are not limited to the examples described above.

[0040] An artificial intelligence model may be composed of multiple neural network layers. Each of the multiple neural network layers has multiple weight values ​​and performs neural network operations through operations between the results of previous layers and the multiple weights. The multiple weights possessed by the multiple neural network layers can be optimized based on the learning results of the artificial intelligence model. For example, the multiple weights may be updated so that the loss value or cost value obtained from the artificial intelligence model during the learning process is reduced or minimized. Artificial neural networks may include deep neural networks (DNNs), such as Convolutional Neural Networks (CNNs), Deep Neural Networks (DNNs), Recurrent Neural Networks (RNNs), Restricted Boltzmann Machines (RBMs), Deep Belief Networks (DBNs), Bidirectional Recurrent Deep Neural Networks (BRDNNs), or Deep Q-Networks, but are not limited to the examples mentioned above.

[0041] According to an exemplary embodiment of the present disclosure, a processor can implement artificial intelligence. Artificial intelligence refers to a machine learning method based on an artificial neural network that enables a machine to learn by mimicking human biological neurons. Methodologies of artificial intelligence can be classified according to the learning method into supervised learning, where input and output data are provided together as training data and the solution (output data) to the problem (input data) is predetermined; unsupervised learning, where only input data is provided without output data and the solution (output data) to the problem (input data) is not predetermined; and reinforcement learning, where a reward is given from an external environment whenever an action is taken from the current state, and learning proceeds in a direction that maximizes such reward. In addition, artificial intelligence methodologies can be classified according to the architecture, which is the structure of the learning model. The architectures of widely used deep learning technologies can be classified into Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN), Transformers, and Generative Adversarial Networks (GAN).

[0042] The device and system may include an artificial intelligence model. The artificial intelligence model may be a single model or may be implemented as multiple models. The artificial intelligence model may be composed of a neural network (or artificial neural network) and may include statistical learning algorithms that mimic biological neurons in machine learning and cognitive science. A neural network may refer to a model that possesses problem-solving capabilities by having artificial neurons (nodes) that form a network through synaptic connections and change the strength of synaptic connections through learning. The neurons of a neural network may include combinations of weights or biases. A neural network may include one or more layers composed of one or more neurons or nodes. For example, the device may include an input layer, a hidden layer, and an output layer. The neural network constituting the device can infer a result (output) to be predicted from an arbitrary input by changing the weights of the neurons through learning.

[0043] The processor can create neural networks, train or learn neural networks, perform computations based on received input data, generate information signals based on the results of the computation, or retrain neural networks. Neural network models may include, but are not limited to, various types of models such as Convolutional Neural Networks (CNN), Region with Convolutional Neural Networks (R-CNN), Region Proposal Networks (RPN), Recurrent Neural Networks (RNN), Stacking-based Deep Neural Networks (S-DNN), State-Space Dynamic Neural Networks (S-SDNN), Deconvolution Networks, Deep Belief Networks (DBN), Restructured Boltzmann Machines (RBM), Fully Convolutional Networks, Long Short-Term Memory Networks (LSTM), and Classification Networks, such as GoogleNet, AlexNet, and VGG Network. The processor may include one or more processors to perform computations according to neural network models. For example, a neural network is a deep neural network It may include a (Deep Neural Network).

[0044] Neural networks include CNN (Convolutional Neural Network), RNN (Recurrent Neural Network), perceptron, multilayer perceptron, FF (Feed Forward), RBF (Radical Basis Function), DFF (Deep Feed Forward), LSTM (Long Short Term Memory), GRU (Gated Recurrent Unit), AE (Auto Encoder), VAE (Variational Auto) Encoder), DAE (Denoising Auto Encoder), SAE (Sparse Auto Encoder), MC (Markov Chain), HN (Hopfield Network), BM (Boltzmann Machine), RBM (Restricted Boltzmann Machine), DBN (Deep Belief Network), DCN (Deep Convolutional Network), DN (Deconvolutional Network), DCIGN (Deep Convolutional Inverse Graphics Network), GAN (Generative Adversarial Network), LSM (Liquid State Machine), ELM (Extreme Learning Machine), ESN (Echo A person skilled in the art will understand that any neural network may be included, but is not limited to, State Network, Deep Residual Network, Differential Neural Computer, Neural Turing Machine, Capsule Network, Kohonen Network, and Attention Network.

[0045] According to an exemplary embodiment of the present disclosure, the processor comprises a Convolutional Neural Network (CNN) such as GoogleNet, AlexNet, VGG Network, Region with Convolutional Neural Network (R-CNN), Region Proposal Network (RPN), Recurrent Neural Network (RNN), Stacking-based Deep Neural Network (S-DNN), State-Space Dynamic Neural Network (S-SDNN), Deconvolution Network, Deep Belief Network (DBN), Restricted Boltzmann Machine (RBM), Fully Convolutional Network, Long Short-Term Memory (LSTM) Network, Classification Network, Generative Modeling, eXplainable AI, Continual AI, Representation Learning, AI for Material Design, BERT, SP-BERT, MRC / QA, Text Analysis, Dialog System, GPT-3, GPT-4 for Natural Language Processing, Visual Analytics, Visual Understanding, Video Synthesis for Vision Processing, Anomaly Detection, Prediction, Time-Series Forecasting, Optimization, Recommendation for ResNet Data Intelligence, Various artificial intelligence structures and algorithms, such as data creation, may be used, but are not limited thereto. Hereinafter, embodiments of the present disclosure will be described in detail with reference to the attached drawings.

[0046] FIG. 1 is a block diagram briefly illustrating the configuration of a device (100) for detecting the formal incompleteness of a self-introduction letter answer according to one embodiment of the present disclosure.

[0047] Referring to FIG. 1, the electronic device (100) according to the present disclosure may include an input / output module (110), a communication module (120), a memory (130), and a processor (140). In the following, the electronic device (100) according to the present disclosure is an electronic device that performs an error category detection operation of a response sentence within a self-introduction letter, and the method according to the present disclosure is assumed to be implemented through the electronic device (100) that performs an error category detection operation of a response sentence within a self-introduction letter.

[0048] The components illustrated in FIG. 1 are not essential for implementing the electronic device (100) according to the present disclosure, so the electronic device (100) described in this specification may have more or fewer components than the components listed above.

[0049] The input / output module (110) may be various interfaces or connection ports that receive input from a user or output information to a user. The input / output module (110) may be divided into an input module and an output module.

[0050] The input module is for inputting video information (or signal), audio information (or signal), data, or information input from a user, and may include at least one of at least one camera, at least one microphone, and a user input unit. Voice data or image data collected by the input module may be analyzed and processed into a user control command.

[0051] User input can take various forms, including key input, touch input, and voice input. Examples of input modules capable of receiving such user input include traditional keypads, keyboards, and mice; as well as touch sensors that detect user touch; microphones that receive voice signals; cameras that recognize gestures through image recognition; proximity sensors consisting of light or infrared sensors that detect user approach; motion sensors that recognize user movements using accelerometers or gyroscopes; and all other diverse forms of input means that detect or receive various types of user input. This is a comprehensive concept.

[0052] The output module can output various types of information and provide it to the user. The output module is intended to generate outputs related to sight, hearing, or touch, and may include at least one of a display unit, an audio output unit, a haptic module, and an optical output unit. In addition, as a comprehensive concept that includes all various forms of output means, it may be implemented in the form of a port-type output interface that connects the individual output means described above.

[0053] The display unit can implement a touch screen by forming a layered structure with the touch sensor or by being formed as an integral unit. Such a touch screen functions as a user input unit that provides an input interface between the device and the user, and at the same time can provide an output interface between the device and the user.

[0054] The display unit displays (outputs) information processed by the device. For example, the display unit may display execution screen information of an application program (e.g., an application) running on the device, or UI (User Interface) and GUI (Graphic User Interface) information based on such execution screen information.

[0055] The interface serves as a passage for various types of external devices connected to the device. This interface section may include at least one of a wired / wireless headset port, an external charger port, a wired / wireless data port, a memory card port, a port for connecting a device equipped with a SIM card, an audio I / O (Input / Output) port, a video I / O (Input / Output) port, and an earphone port. The device can perform appropriate control related to the external device connected to the interface section.

[0056] In other words, the input / output module (110) can receive user input or provide output to the user based on a user interface.

[0057] Among the above components, the communication module (120) may include one or more components that enable communication with an external device, and may include, for example, at least one of a broadcast receiving module, a wired communication module, a wireless communication module, a short-range communication module, and a location information module.

[0058] In addition to Wi-Fi modules and WiBro (Wireless broadband) modules, the wireless communication module may include wireless communication modules that support various wireless communication methods such as GSM (global System for Mobile Communication), CDMA (Code Division Multiple Access), WCDMA (Wideband Code Division Multiple Access), UMTS (universal mobile telecommunications system), TDMA (Time Division Multiple Access), LTE (Long Term Evolution), 4G, 5G, and 6G.

[0059] The wireless communication module may include a wireless communication interface comprising an antenna and a transmitter that transmit a mobile communication signal. Additionally, the wireless communication module may further include a signal conversion module that modulates a digital control signal output from the processor through the wireless communication interface into an analog wireless signal under the control of the processor.

[0060] The wireless communication module may include a wireless communication interface comprising an antenna and a receiver for receiving the signal. Additionally, the wireless communication module may further include a signal conversion module for demodulating an analog wireless signal received through the wireless communication interface into a digital control signal.

[0061] A short-range communication module is for short-range communication and can support short-range communication by using at least one of Bluetooth™, RFID (Radio Frequency Identification), Infrared Data Association (IrDA), UWB (Ultra Wideband), ZigBee, NFC (Near Field Communication), Wi-Fi (Wireless-Fidelity), Wi-Fi Direct, and Wireless USB (Wireless Universal Serial Bus) technologies.

[0062] The memory (130) can store data supporting various functions of the device and programs for the operation of the processor, and can store input / output data (e.g., music files, still images, videos, etc.), and can store a number of application programs (or applications) running on the device, data for the operation of the device, and instructions. At least some of these application programs can be downloaded from an external server via wireless communication.

[0063] Such memory may include at least one type of storage medium among flash memory type, hard disk type, SSD type (Solid State Disk type), SSD type (Silicon Disk Drive type), multimedia card micro type, card type memory (e.g., SD or XD memory, etc.), RAM (random access memory; RAM), SRAM (static random access memory), ROM (read-only memory; ROM), EEPROM (electrically erasable programmable read-only memory), PROM (programmable read-only memory), magnetic memory, magnetic disk, and optical disk. Additionally, the memory may be a database that is separate from the device but connected via wired or wireless connection.

[0064] The processor (140) may be implemented as a memory that stores data for an algorithm or a program that reproduces the algorithm for controlling the operation of components within the device, and at least one processor that performs the aforementioned operation using the data stored in the memory. In this case, the memory (130) and the processor (140) may each be implemented as separate chips. Alternatively, the memory (130) and the processor (140) may be implemented as a single chip.

[0065] The processor (140) may be implemented as a computer or a similar device according to hardware, software, or a combination thereof. Hardware-wise, the processor (140) may be provided in the form of an electronic circuit that processes electrical signals to perform control functions, and software-wise, it may be provided in the form of a program that drives the hardware processor. Meanwhile, unless otherwise specifically mentioned in the following description, the operation of the first device and / or the second device may be interpreted as being performed by the control of the processor (140). That is, the modules may be interpreted as the processor (140) controlling the first device and / or the second device to perform the following operations.

[0066] In addition, the processor (140) can control one or a combination of the components described above in order to implement various embodiments according to the present disclosure described in FIGS. 2 to 8 below on the device.

[0067] At least one component may be added or removed in response to the performance of the components illustrated in FIG. 1. Additionally, it will be readily understood by those skilled in the art that the relative positions of the components may be changed in response to the performance or structure of the system.

[0068] Meanwhile, each component illustrated in Fig. 1 refers to a software and / or hardware component such as a Field Programmable Gate Array (FPGA) and an Application Specific Integrated Circuit (ASIC).

[0069] FIG. 2 is a block diagram briefly illustrating the process of a device (100) for detecting the morphological incompleteness of a self-introduction letter answer according to one embodiment of the present disclosure. FIG. 3 is an algorithm (100) illustrating the Request process performed by the device (100) for detecting the morphological incompleteness of a self-introduction letter answer according to one embodiment of the present disclosure, FIG. 4 is an algorithm illustrating the Response process. FIG. 5 is a flowchart illustrating the non-answer category output process of the device (100) for detecting the morphological incompleteness of a self-introduction letter answer according to one embodiment of the present disclosure, FIG. 6 is a flowchart illustrating the non-word category output process, FIG. 7 is a flowchart illustrating the non-sentence category output process.

[0070] Hereinafter, with reference to FIGS. 2 to 7, a device (100) for detecting the formal incompleteness of a self-introduction letter response according to the present disclosure will be described.

[0071] The answer sentences entered for each question in recruitment documents, particularly in the personal statement, serve as recruitment data necessary for hiring and must be extracted under the following premises: the answer sentences in the personal statement must be in the form of answers, written as sentences, and each sentence must consist of individual words.

[0072] In other words, the answer sentences entered for each question item must possess morphological completeness and must not be non-sentences, non-words, or non-answers. The morphological incompleteness of the answer sentences may include non-sentences, non-words, and non-answers as detailed categories, and may include bullets classified as a bulleted narrative form as an exceptional sentence writing style.

[0073] Therefore, answer sentences that are not non-answers, non-sentences, or non-words can be extracted and utilized as recruitment data.

[0074] At least one processor (140) of the device (100) for detecting the formal incompleteness of a self-introduction letter response according to the present disclosure may be configured to obtain a response sentence from the input self-introduction letter, detect an error category corresponding to each of the response sentences among predefined error categories, list at least one error category detected in each response sentence, and provide each response sentence and the listed error category to the input / output module.

[0075] In one embodiment, the error category may include at least one of a non-answer that is not in the form of an answer, a non-sentence that is not in the form of a sentence structure, a non-word that is not a word, a minor error containing punctuation that does not conform to the rules, and a bullet that is a sentence written in a bulleted descriptive form.

[0076] In other words, defects that may occur in response sentences can be defined as non-responses, non-sentences, and non-words; additionally, minor sentence errors can be defined as minor errors, and bulleted forms of narrative where the ending is a noun, differing from general sentence structure, can be defined as bullets.

[0077] As an example, one answer sentence may correspond to at least one of a non-answer, a non-sentence, or a non-word.

[0078] If a single answer sentence corresponds to both non-sentences and non-words, it can be classified as a non-sentence category.

[0079] Alternatively, if a single answer sentence contains parts that are non-words and parts that are non-sentences, and the non-word and non-sentence parts overlap only partially but are not entirely identical, it can be classified into both the non-word category and the non-sentence category.

[0080] Specifically, as illustrated in FIG. 2, at least one processor (140) of an electronic device (100) according to one embodiment of the present disclosure may perform a step of obtaining an answer sentence (S210), a step of defining an error category (S220), a step of classifying an error category (S230), and a step of providing (S240).

[0081] In the step of obtaining an answer sentence (S210), at least one processor (140) of the electronic device (100) according to one embodiment of the present disclosure may be configured to obtain the answer sentence by deleting a pre-set question sentence within the self-introduction letter and extracting the remaining sentence.

[0082] Since the question sentences in recruitment documents, such as a self-introduction letter, may vary from company to company, the question sentences within the self-introduction letter can be removed and the answer sentences obtained based on the company name stored in the database and the question sentences of the said company.

[0083] In the error category definition step (S220), error types corresponding to the error categories can be defined. Specifically, for at least one of the following: non-answers that are not in the form of an answer, non-sentences that are not in the form of a sentence, non-words that are not words, minor errors containing punctuation that does not conform to rules, and bullets that are sentences written in a bulleted form, it can be defined what the answer sentence corresponding to each type is.

[0084] In the error category classification step (S230), the obtained answer sentence can be classified as belonging to which category among the types defined in the error category definition step (S220).

[0085] As an example, if the last sentence of the answer includes "please," "please," or "explain," it can be classified as a non-answer.

[0086] As an example, a string of characters that continues without spaces or a string of characters composed mainly of consonants or vowels can be classified as a non-word.

[0087] As an example, if the last word of the answer sentence is a string that does not end with a predicate and is not a modified sentence, it can be classified as a non-sentence.

[0088] As an example, if there is no period at the end of the answer sentence, a comma is used instead of a period, or there is one or more periods, it can be classified as a minor error as a punctuation error.

[0089] As an example, if the last word of the answer sentence is a pre-set bulleted language that is frequently used in bulleted descriptions, it can be classified as a bulleted sentence.

[0090] In the provision step (S240), an answer sentence and at least one error category corresponding to the answer sentence can be mapped and provided to the user interface module. In other words, it can be provided so that the user can identify and monitor the answer sentence and the error type corresponding to the answer sentence.

[0091] Specifically, as illustrated in FIGS. 3 and 4, at least one processor (140) of an electronic device (100) according to one embodiment of the present disclosure can perform a Response algorithm (40) according to a Request algorithm (30) for an acquired answer sentence.

[0092] Referring to Fig. 3, it can be seen that answer sentences are extracted for each question item, and the answer sentences are separated into sentences to obtain sentences.

[0093] In this regard, detect_list can be used to define a separate list of defects to check for each statement.

[0094] For example, regarding “Making people visit the bank continuously is fulfilling one’s responsibilities as an employee. Specific experience” and “2015.01~2018.03 Planning of HR system improvement and employee training programs,” you can request detection for the presence of nonsent, bullet, nonword, and nonansw errors.

[0095] Additionally, regarding “(Include description if relevant work experience exists among financial specialists or global applicants)”, “2015.01~2018.03 Planning of HR system improvement and employee training programs”, “The club was able to survive because of a dedicated heart”, and “Before applying to the Korea Health and Welfare Human Resource Development Institute, I spent time with the elderly and children doing rural work and simple education”, you may request detection for non-sentence, non-word, or non-answer errors.

[0096] Referring to Figure 4, it can be seen that each error type is defined as bullet, non-answer, non-sentence, and non-word, and sentence is organized into a single value to detect error categories.

[0097] As an example, the value can be obtained by removing duplicate sentences from the sentence and preprocessing only the part of the sentence corresponding to the error category instead of taking the entire sentence as is.

[0098] For example, “2015.01~2018.03 Planning of HR system improvement and employee training programs” is an answer sentence that is answered redundantly to two question items, so one can be removed and only the other can be obtained.

[0099] In addition, in “Because there was a dedicated heart, the club could be revived,” “dedicated” is a word located 6 spaces after “was,” which is the last word used to check for non-grammatical sentences; by removing the word after the 6th space, the answer sentence can be obtained. This is an exemplary figure and the value may change.

[0100] In addition, the answer sentence “ㅇㅁ누리'ㅏㅁㄴㅇ루;ㅏㅁㄴㅇ루;ㅣㅁㄴ으;룸ㄴ;” can be obtained by extracting only the strings listed without spaces and exceeding the preset length from “Before applying to the Korea Health and Welfare Human Resource Development Institute, I spent time with the elderly and children doing rural work and simple education.”

[0101] Then, using body_index, sentence_index, begin_offset, and end_offset, you can obtain data on whether an error was detected in the answer sentence of a question item, and on the answer sentences where a specific error category was detected.

[0102] As an example, additional data regarding error category-related supplementary information can be obtained using value2 and value3.

[0103] For example, in an answer sentence separated by morphemes, the type of each morpheme can be mapped.

[0104] In addition, only answer sentences separated into bullet points can be extracted and saved separately.

[0105] Accordingly, at least one processor (140) of the electronic device (100) according to one embodiment of the present disclosure must specifically define the characteristics of the answer sentence corresponding to the error category.

[0106] According to one embodiment, the at least one processor (140) may be configured to divide the answer sentence into words based on spaces, and if the last word of the answer sentence is a term stored in a dictionary, or if the last word of the answer sentence is a first term stored in a dictionary and a second term stored in a dictionary is placed consecutively with the first term or with one word in between, the answer sentence may be detected as a non-answer among the error categories.

[0107] As illustrated in FIG. 5, at least one processor (140) can separate the acquired answer sentences by word (S510). Before separating by word, preprocessing can be performed to remove duplicate answer sentences from the answer sentences or to remove strings placed at a position that is a preset distance from the last word of the sentence.

[0108] Answer sentences can be separated by spaces to distinguish them by word.

[0109] For words separated by word segments, check whether it is the last word (S520), and if it is the last word (YES in S520), check whether the last word is a term stored in the dictionary (S530).

[0110] Here, the term "word" can include not only words that can be separated and used independently in the dictionary sense, but also combinations of words.

[0111] As an example, a term stored in a dictionary may be a common word used as the last word of a sentence in a question item. For example, a term stored in a dictionary may be "please," "please," "explain," etc.

[0112] As another embodiment, the terms stored in advance may be the last word of a sentence in a question item and the word before the last word, or the last word, the first word before the last word, and the second word before the last word.

[0113] For example, the second word prior to the last word above may be 'write, write, describe, describe', the first word prior to the last word above may be 'please', and the last word above may be 'please'. Or, the first word prior to the last word above may be 'describe, describe', and the last word above may be 'please'.

[0114] At this time, the last word of the above answer sentence is a first term stored in a dictionary, and the second term stored in a dictionary may mean the first word preceding the last word or the second word preceding the last word, which is placed consecutively with the first term or with one word in between.

[0115] Therefore, if the last word corresponds to a term stored in the above dictionary (YES in S530), the above answer sentence can be output as a non-answer category (S540).

[0116] Additionally, although not illustrated in the drawings, as an example, if 'description including' or 'within Bytes' is included among the words separated by word segments, even if it is not the last word of the answer sentence (NO in S520), it can be determined as a non-answer.

[0117] In other words, the electronic device (100) according to the present disclosure can build a separate database to store question items within a company-specific self-introduction letter, and then remove sentences identical to the question items and extract answer sentences.

[0118] Alternatively, the electronic device (100) according to the present disclosure may compare the last word of an answer sentence with a dictionary-stored term such as 'please, please, explain', and if it corresponds to a dictionary-stored term, it may be removed as a question item and other answer sentences may be extracted.

[0119] In this case, extraction can be performed in the same way even if a period or stop is added to the last word.

[0120] Additionally, the electronic device (100) according to the present disclosure may remove strings that can generally be included in question items, such as 'includes description' and 'within Bytes', from the answer sentence if they are included in the answer sentence, even if they are not the last word.

[0121] Meanwhile, at least one processor (140) of an electronic device (100) according to one embodiment of the present disclosure may be configured to detect the answer sentence as a non-word among the error categories if the answer sentence includes a string of characters that is continuous for a length greater than or equal to a preset length without spaces, or if the answer sentence includes a string in which only consonants or vowels are listed.

[0122] As shown in FIG. 6, if the answer sentence corresponds to a string of length greater than or equal to a preset length without spaces (YES in S610), or if it does not correspond to the first pattern or the second pattern (NO in S640), the answer sentence can be output as a non-word category (S630).

[0123] If it corresponds to the first pattern or the second pattern (YES in S640), the above answer sentence can be excluded from the non-word category and terminated.

[0124] As an example, the preset length may mean a case where 30 or more characters are written consecutively, and a value that is considered difficult to write consecutively without spacing according to general Korean grammar may be arbitrarily set.

[0125] As an example, the first pattern may refer to a pattern in which a specific number of digits and periods are repeated in the string. For example, if it is a pattern in which four digits and periods, two digits and periods, and two digits are listed, or a pattern in which two digits and periods, two digits and periods, and two digits are listed, or a pattern in which two digits and periods and two digits are listed, it may be determined to be a date and the answer sentence may be excluded from the non-word category.

[0126] Additionally, the second pattern may refer to a pattern in which the string consists of an English letter, a number, a period, or a slash following a specific English string. For example, if an English letter, a number, a period (.), or a slash ( / ) is included after “https: / / ”, it may be determined to be a URL address and configured to exclude the answer sentence from the non-word category.

[0127] In one embodiment, an electronic device (100) according to one embodiment of the present disclosure may exclude a response sentence having the first pattern and the second pattern from a non-word category if the string is less than or equal to a preset number. For example, if the string is less than 50 characters, a first pattern in which numbers and periods are repeated may be determined to be a date, and if the string is less than 60 characters, a second pattern including English letters, numbers, periods, or diagonals after a specific English string may be determined to be a URL address.

[0128] In other words, the maximum string length of the first pattern and the maximum string length of the second pattern can be set differently.

[0129] In addition, even if the answer sentence is not a string longer than a preset length without spaces (NO in S610), if consonants or vowels are listed individually (YES in S620), it can be output as a non-word category (S630).

[0130] For example, when a sentence is broken down into individual characters, it may refer to cases where consonants or vowels such as ㅁ, ㄷ, ㅏ, and ㅓ are written alone, rather than words.

[0131] As an example, if the distance between a consonant or a vowel is less than a specific distance, the area can be expanded so that a sentence containing a consonant or vowel within the specific distance is viewed as a single answer sentence and output as a non-word category.

[0132] For example, if it is defined that the sentence area is expanded when the distance is less than 5, in “ㅁㄷ롬ㅇ롬” the distance between ㅁ and ㄷ is 1 and the distance between ㄷ and ㅇ is 2, so “ㅁㄷ롬ㅇ롬” can be viewed as a single sentence and the answer sentence can be output as a non-word category.

[0133] As an example, if the distance between consonants or vowels is greater than a certain distance, the area can be separated into multiple answer sentences and each can be output as a non-word category.

[0134] For example, if it is defined that the sentence area is expanded when the distance is less than 5, then in “I am applying for the second half of the youth internship”, the distance between ㅁㄷㅇㄹ and ㄹㄹㄹㄹㄹ corresponds to 5 or more, so “ㅁㄷㅇㄹ” and “ㄹㄹㄹㄹㄹ” can be output as two non-word categories each.

[0135] Accordingly, at least one processor (140) according to one embodiment of the present disclosure may determine that it does not belong to a non-word category if it is not a string of length greater than or equal to a preset length without spaces (NO in S610) and is not a sequence of consonants or characters alone (NO in S620).

[0136] Meanwhile, at least one processor (140) according to one embodiment of the present disclosure may be configured to detect the answer sentence as a non-sentence among the error categories if the last word of the answer sentence is a word that is not a predicate and the last word of the answer sentence is not a form in which a common noun of a predefined modal descriptive form exists alone.

[0137] As an example of one embodiment, as illustrated in FIG. 7, at least one processor (140) according to one embodiment of the present disclosure can separate a response sentence by morpheme (S710). A response sentence can be separated by morpheme using a natural language processing module stored in an electronic device (100) according to one embodiment of the present disclosure or a natural language processing module stored in a separate server.

[0138] A morpheme is the smallest linguistic unit that carries meaning, and the natural language processing module may include a morpheme analyzer. The morpheme analyzer may be a generally known morpheme analyzer.

[0139] Then, if the morpheme processed by at least one processor (140) according to one embodiment of the present disclosure is the last morpheme of the answer sentence (YES in S720) and the last morpheme is a morpheme corresponding to a predicate (YES in S730), a minor error category can be output for the answer sentence (S790).

[0140] The answer sentence must end with a period; therefore, since the last morpheme must be a punctuation mark, a minor error category may be output as a punctuation error.

[0141] Additionally, at least one processor (140) according to one embodiment of the present disclosure may be configured to detect that the answer sentence corresponds to a minor error among the error categories if the last morpheme of the answer sentence is not another symbol corresponding to at least one of a period, an exclamation mark, and a question mark.

[0142] Even if the last morpheme corresponds to a punctuation mark, if it is not at least one of a period, exclamation mark, or question mark, or if multiple periods, exclamation marks, or question marks are present, the above answer sentence may be output as a minor error category.

[0143] Alternatively, if the morpheme processed by at least one processor (140) according to one embodiment of the present disclosure is the last morpheme of the answer sentence (YES in S720), and the last morpheme is not a morpheme corresponding to a predicate (NO in S730), and the last morpheme is a common noun in the form of a bulleted predicate (YES in S740), a bullet category can be output for the answer sentence.

[0144] If the above last morpheme is a common noun and corresponds to a common noun in the predefined bulleted descriptive form, it may be classified as a bulleted sentence.

[0145] For example, an electronic device (100) according to one embodiment of the present disclosure includes 'improvement', 'management', 'evaluation', 'contribution', 'progress', 'possibility', 'reinforcement', 'creation', 'work', 'contribution', 'increase', 'experience', 'improvement', 'guidance', 'understanding', 'development', 'support', 'reappointment', 'research', 'selection', 'responsibility', 'issuance', 'verification', 'configuration', 'expression', 'service', 'acquisition', 'consulting', 'completion', 'selection', 'provision', 'achievement', 'record', 'accumulation', 'identification', 'saving', 'assistance', 'application', 'promotion', 'utilization', 'enhancement', 'cultivation', 'award', 'guidance', 'mindset', 'planning', 'inspection', 'activity', At least one of the words 'acquisition', 'increase', 'ability', 'effort', 'cultivation', 'acquisition', 'seeking', 'displaying', 'rise', 'processing', 'promotion', 'proposal', 'persuasion', and 'challenge' can be stored in the dictionary as a common noun in a bulleted descriptive form.

[0146] Alternatively, if the morpheme processed by at least one processor (140) according to one embodiment of the present disclosure is the last morpheme of the answer sentence (YES in S720), and the last morpheme is not a morpheme corresponding to a predicate (NO in S730), and the last morpheme is not a common noun in the form of a modified predicate (NO in S740), a non-sentence category can be output for the answer sentence.

[0147] In one embodiment, at least one processor (140) according to one embodiment of the present disclosure can output a non-sentence category for the answer sentence by confirming that the last morpheme is not a morpheme corresponding to a punctuation mark, and then confirming that the last morpheme is not a morpheme corresponding to a predicate and a common noun in a modified descriptive form.

[0148] Alternatively, as an example, at least one processor (140) according to an example of the present disclosure may output a non-sentence category for the answer sentence by confirming that the previous morpheme is not a morpheme corresponding to a predicate and a common noun in a modified descriptive form when the last morpheme is a morpheme corresponding to a punctuation mark.

[0149] Additionally, at least one processor (140) according to one embodiment of the present disclosure may be further configured to detect that the answer sentence belongs to the non-sentence category if the morpheme other than the last morpheme of the answer sentence corresponds to a pre-final ending (YES at S720 and YES at S760), and the morpheme following the pre-final ending does not correspond to at least one of a terminal ending, a connecting ending, a derivational ending, a pre-final ending, a nominal derivational ending, an adjective derivational ending, an adjective derivational suffix, and an accusative case particle (NO at S770).

[0150] In one embodiment, among the morphemes included in the answer sentence, if a terminal ending (EF), a connecting ending (EC), a derivational ending (ET), another pre-final ending (EP), a nominal derivational ending (ETN), an adjectival derivational ending (ETM), or an adjective derivational suffix (XSA) is placed at a position following a pre-final ending (EP), or if an adjective derivational suffix (XSA) and an adjectival derivational ending (ETM) or a nominal derivational ending (ETN) and an accusative case particle (JKO) are not located, the answer sentence can be detected as belonging to a non-sentence category.

[0151] For example, in the response sentence “I performed response duties based on trust with customers” processed by at least one processor (140), the morpheme other than the last morpheme includes the pre-final ending (EP) “겠”, and the next morpheme is ‘저’, which is a pronoun morpheme (NP) and does not correspond to a pre-set type of morpheme, so the above response sentence can be detected as a non-sentence category.

[0152] Meanwhile, according to one embodiment of the present disclosure, the at least one processor (140) may be configured to output that the answer sentence corresponds to a bullet among the error categories (S780) if the last word of the answer sentence is a word that exists alone as a common noun in a predefined bulleted descriptive form (YES in S740), and to provide the answer sentence to the input / output module (110) to receive input regarding whether the answer sentence is a bullet or a non-sentence.

[0153] Since bullet categories and non-sentence categories are highly likely to be incorrectly displayed, the above answer sentence can be displayed in a bullet category, but additional information that allows for feedback can be provided along with it.

[0154] Meanwhile, the non-sentence category, in which the last morpheme of a sentence is a morpheme that does not correspond to a predicate and is not a common noun in the form of a bulleted predicate, and the minor error category, in which the last morpheme of a sentence is a morpheme that does not correspond to a punctuation mark, have similar parts and must be accurately defined so that at least one processor (140) can make a judgment.

[0155] In one embodiment, if the last morpheme is at least one of a conjunction (EC), a positive modifier (VCP) and a conjunction (EC), an adjectival derivational ending (ETM), a positive modifier (VCP) and an adjectival derivational ending (ETM), and the last word corresponds to at least one of go, ni, seumnida, doro, yeo, jiman, myeo, ji, eoyong, ya, neun, in, ge, gi, then at least one processor (140) can detect the answer sentence as a non-sentence category.

[0156] For example, in “growing as a corporate finance expert,” the last morpheme is a connecting ending (EC) and the last word is “go,” so it may fall into the non-sentence category.

[0157] In other cases, at least one processor (140) can detect the answer sentence as a minor error category.

[0158] In one embodiment, if the last morpheme is not a modified descriptive form but is at least one of a common noun (NNG), an accusative case particle (JKO), a proper noun (NNP), an auxiliary particle (JX), and a conjunction particle (JC), and the last word corresponds to at least one of time, thought, of, etc., object, habit, or thing, then at least one processor (140) can detect the answer sentence as a non-sentence category.

[0159] For example, in “evaluation that acts as a positive catalyst”, the last morpheme is an accusative case particle and the last word corresponds to ‘를’, so the above answer sentence can be detected as a non-sentence category.

[0160] In one embodiment, if the last morpheme corresponds to at least one of a verb derivation suffix (XSV), an adverbial particle (JKB), a verb derivation suffix (XSV) and an adjectival derivation ending (ETM), a nominative particle (JKS), a foreign language (SL), a dependent noun (NNB), an adjective derivation suffix (XSA) and an adjectival derivation ending (ETM), a verb (VV) and an adjectival derivation ending (ETM), an adjective (VA) and an adjectival derivation ending (ETM), and an adjective (VA) and an adjective derivation suffix (XSA), the answer sentence can be detected as a non-sentence category.

[0161] For example, in “an employee contributing to performance,” the last morphemes, employee (NNG) and this (JKS), correspond to nominative case particles and can be detected as a non-sentence category.

[0162] In one embodiment, if the last morpheme corresponds to at least one of a terminal ending (EF), a verb derivation suffix (XSV) and a connecting ending (EC), an adjective derivation suffix (XSA) and a connecting ending (EC), a verb (VV) and a terminal ending (EF), an auxiliary predicate (VX) and a connecting ending (EC), a pre-final ending (EP) and a connecting ending (EC), a positive demonstrator (VCP) and a terminal ending (EF), a verb (VV) and a pre-final ending (EP) and a terminal ending (EF), a negative demonstrator (VCN) and a connecting ending (EC), a verb (VV) and a connecting ending (EC), and a verb derivation suffix (XSV) and a terminal ending (EF), the last word can be detected as a minor error category if it corresponds to one of 습니다, 합니다, 한다, 되니다, 십니다, 예요, 사세요, 못습니다, 아니니다, 시키다.

[0163] In addition, if the last word is 'haya', it can be detected as a non-sentence category.

[0164] In other cases, due to an undefined defect, at least one processor (140) may provide the above answer sentence to the input / output module (110).

[0165] For example, in “must proceed with proactive work,” 'might' can be detected as a non-sentence category as a morpheme composed of a verb-deriving suffix and a connective ending.

[0166] In one embodiment, if the last morpheme corresponds to at least one of a number (SN), a general adverb (MAG), an adjective (MM), an interjection (IC), a root (XR), an auxiliary verb (VX), and a common noun (NNG), it can be detected as a non-sentence category.

[0167] In one embodiment, if the answer sentence is composed of two or more morphemes, the last morpheme is a separator (SC), and the morpheme preceding the last morpheme is at least one of a terminal ending (EF), a verb derivation suffix (XSV) and a terminal ending (EF), a verb (VV) and a pre-terminal ending (EP) and a terminal ending (EF), an adjective derivation suffix (XSA) and a terminal ending (EF), a verb (VV) and a terminal ending (EF), a period, a question mark, an exclamation mark (SF), a positive modifier (VCP) and a terminal ending (EF), a negative modifier (VCN) and a terminal ending (EF), the answer sentence may be detected as a minor error category.

[0168] In other cases, it may be detected as a non-sentence category.

[0169] Additionally, at least one processor (140) can detect non-defective answer sentences using a pre-stored exception handling algorithm.

[0170] As an example, answer sentences whose last words are 'ga.', 'na.', or 'da.' can be pre-stored in an exception handling algorithm. This is because these are terms frequently used for sentence separation and may serve as expressions to distinguish the end of a sentence from the beginning of the next sentence.

[0171] In one embodiment, if the first morpheme is an opening bracket (SSO) and the answer sentence containing the first morpheme is not the last sentence, the answer sentence may be pre-stored by an exception handling algorithm.

[0172] As an example, if the last morpheme is a closing parenthesis (SSC) and the string preceding the last morpheme has a first pattern corresponding to weather, the answer sentence can be pre-stored as an exception handling algorithm.

[0173] In addition, the answer sentence can belong to multiple categories.

[0174] According to one embodiment of the present disclosure, at least one processor (140) may output only the non-sentence category if the entire sentence of the answer sentence corresponds to both the non-sentence category and the non-word category.

[0175] For example, the answer sentence “I am applying for the second half of the year youth internship” belongs to the non-word category due to “I am applying for the second half of the year youth internship” and may also belong to the non-sentence category due to “I am applying for the second half of the year youth internship”. In this case, at least one processor (140) can output only the non-sentence category. This is to improve preprocessing efficiency, as sentences belonging to the non-sentence category are more likely to be longer.

[0176] Alternatively, at least one processor (140) according to one embodiment of the present disclosure may be configured to output an error category corresponding to the longest sentence among the lengths of sentences corresponding to each error category when the answer sentence corresponds to a plurality of error categories.

[0177] It is possible to distinguish error categories and corresponding answer sentences from the very beginning and detect error categories based on the length of each answer sentence.

[0178] FIG. 8 is a flowchart illustrating a detection method of a device (100) for detecting the morphological incompleteness of a self-introduction letter response according to one embodiment of the present disclosure.

[0179] As illustrated in FIG. 8, a processing method of a device (100) for detecting the formal incompleteness of a self-introduction letter response according to one embodiment of the present disclosure may include the steps of: obtaining a response sentence from the input self-introduction letter (S810); detecting an error category corresponding to each of the response sentences among a predefined error category (S820); listing at least one error category detected in each response sentence (S830); and providing each response sentence and the listed error category to an input / output module (110) (S840).

[0180] Content that overlaps with the above is omitted for the sake of brevity in the specification.

[0181] Meanwhile, the disclosed embodiments may be implemented in the form of a recording medium that stores instructions executable by a computer. The instructions may be stored in the form of program code and, when executed by a processor, may generate a program module to perform the operation of the disclosed embodiments. The recording medium may be implemented as a computer-readable recording medium.

[0182] Computer-readable recording media include all types of recording media that store instructions that can be decoded by a computer. Examples include ROM (Read Only Memory), RAM (Random Access Memory), magnetic tape, magnetic disk, flash memory, optical data storage devices, etc.

[0183] As described above, the disclosed embodiments have been explained with reference to the attached drawings. Those skilled in the art will understand that the present disclosure may be practiced in forms different from the disclosed embodiments without changing the technical spirit or essential features of the present disclosure. The disclosed embodiments are illustrative and should not be interpreted restrictively. Explanation of the symbols

[0184] 100: Electronic device 110: I / O module 120: Communication module 130: Memory 140: Processor

Claims

Claim 1 A device for detecting the morphological incompleteness of a self-introduction response, comprising: an input / output module that receives user input or provides output to the user based on a user interface; a memory in which at least one process is stored for performing an operation to detect an error category of a sentence within a self-introduction; and at least one processor that performs an operation to detect an error category of a sentence within a self-introduction according to the process; wherein the at least one processor is configured to acquire a response sentence from the input self-introduction, detect an error category corresponding to each of the response sentences among predefined error categories, list at least one error category detected in each response sentence, and provide each response sentence and the listed error category to the input / output module; wherein the error category includes at least one of a non-answer that is not in the form of an answer, a non-sentence that is not in the form of a sentence, a non-word that is not a word, a minor error containing punctuation that does not conform to rules, and a bullet that is a sentence written in a bulleted descriptive form; and wherein the error category is defined as a criterion for determining whether each response sentence is morphologically complete in order to determine the direction of preprocessing of each response sentence. Claim 2 In claim 1, the device for detecting the morphological incompleteness of a self-introduction answer is configured such that at least one processor deletes a pre-set question sentence within the self-introduction and extracts the remaining sentence to obtain the answer sentence. Claim 3 A device for detecting the morphological incompleteness of a self-introduction response according to claim 2, wherein at least one processor divides the answer sentence into phrases based on spaces, and detects the answer sentence as a non-answer among the error categories if the last word of the answer sentence is a term stored in a dictionary, or if the last word of the answer sentence is a first term stored in a dictionary and a second term stored in a dictionary is included in a phrase that is placed consecutively with the first term or with one word in between. Claim 4 A device for detecting the morphological incompleteness of a self-introduction response, wherein at least one processor is configured to detect the answer sentence as a non-word among the error categories if the answer sentence contains a string of characters of a length greater than or equal to a preset length without spaces, or if the answer sentence contains a string of characters in which only consonants or vowels are listed. Claim 5 A device for detecting the morphological incompleteness of a self-introduction response according to claim 4, wherein at least one processor is configured to exclude the answer sentence from the non-word category if the answer sentence contains a string of consecutive characters of a length greater than or equal to a preset length without spaces, or if the string has a pattern of repeating a specific number of numbers and periods, or if the string consists of a specific English string followed by an English letter, a number, a period, or a diagonal line. Claim 6 A device for detecting the morphological incompleteness of a self-introduction response, wherein, in claim 2, at least one processor is configured to detect the answer sentence as a non-sentence among the error categories if the last word of the answer sentence is a word that is not a predicate and the last word of the answer sentence is not a form in which a common noun of a predefined modal descriptive form exists alone. Claim 7 A device for detecting the morphological incompleteness of a self-introduction letter response according to claim 6, wherein at least one processor processes the answer sentence by natural language and separates it by morpheme, and detects that the answer sentence belongs to the non-sentence category if a morpheme other than the last morpheme of the answer sentence corresponds to a pre-final ending, and the morpheme following the pre-final ending does not correspond to at least one of a terminal ending, a connecting ending, a derivational ending, a pre-final ending, a nominal derivational ending, an adjective derivational ending, an adjective derivational suffix, and an accusative case particle. Claim 8 A device for detecting the morphological incompleteness of a self-introduction response according to claim 6, wherein at least one processor processes the answer sentence by natural language and separates it by morpheme, and detects that the answer sentence corresponds to a minor error among the error categories when the last morpheme of the answer sentence is a word that is a predicate, or when the last morpheme of the answer sentence is not another symbol corresponding to at least one of a period, an exclamation mark, and a question mark. Claim 9 A device for detecting the morphological incompleteness of a self-introduction response according to claim 1, wherein at least one processor outputs that the answer sentence corresponds to a bullet among the error categories if the last word of the answer sentence is a word that exists alone as a common noun in a predefined bulleted descriptive form, provides the answer sentence to the input / output module to receive input regarding whether the answer sentence is a bullet or a non-sentence, and outputs only the non-sentence category if the entire sentence of the answer sentence corresponds to both the non-sentence category and the non-word category, or outputs the error category corresponding to the longest sentence among the lengths of the sentences corresponding to each error category if the answer sentence corresponds to multiple error categories. Claim 10 A method for detecting the morphological incompleteness of a self-introduction response, performed by at least one processor of an electronic device, comprising: obtaining a response sentence from an input self-introduction; detecting an error category corresponding to each of the response sentences among predefined error categories; listing at least one error category detected in each response sentence; and providing each of the response sentences and the listed error categories to an input / output module of the electronic device that receives user input or provides output to the user based on a user interface, wherein the error categories include at least one of a non-answer that is not in the form of an answer, a non-sentence that is not in the form of a sentence, a non-word that is not a word, a minor error containing punctuation that does not conform to rules, and a bullet that is a sentence written in a bulleted descriptive form, wherein the error categories are defined as criteria for determining whether each of the response sentences is morphologically complete in order to determine the preprocessing direction of each of the response sentences.

Images