Entity weight scoring method, system, electronic device and storage medium

Abstract

The present invention proposes an entity weight scoring method, system, electronic device and storage medium, and the method and technical scheme thereof include a standard library establishment step, setting a standard library, setting standard entity attributes, standard emotional states, and weight values ​​of the standard emotional states in the standard library; a data acquisition step, acquiring the original evaluation data of a product on an e-commerce platform; a data preprocessing step, filtering redundant information from the original evaluation data, and performing word segmentation processing; an entity processing step, performing entity level processing on the preprocessed original evaluation data, and establishing an entity emotion correspondence table according to the entity level processing result; and a scoring step, calculating the score of the product according to the entity emotion correspondence table and the standard library. The present invention can solve the problem that the existing entity weight scoring method does not make sufficient use of emotional states and is unscientific.

Description

Technical Field

[0001] The present invention belongs to the field of natural language processing, and in particular relates to an entity weight scoring method, system, electronic device and storage medium. Background Art

[0002] People can express their feelings about the products they buy online and offline on the Internet, and their emotional state is included. By capturing this data, we can obtain the public's feedback on the entity and give a certain score, so that product producers and advertisers can be more targeted in advertising and improving the performance of the entity. However, due to the huge amount of data, it is cumbersome and time-consuming to simply organize the data manually, requiring a lot of manpower and financial resources. At the same time, the subjective emotions of humans in the scoring process make the scoring non-objective.

[0003] Product evaluations on the Internet are complex, and the evaluations of entities and entity attributes involved may be presented in the evaluation data. Most existing technologies collect logs, then make predictions on user-entity analysis and user daily behavior to recommend entities, or analyze evaluation logs to label the evaluation as positive or negative, rather than scoring by the weight of the entity and its attributes to achieve the purpose of targeted product optimization. In addition, entity scoring is often done manually, with subjective scoring, which has the disadvantages of error and long manual time. Summary of the invention

[0004] The embodiments of the present application provide an entity weight scoring method, system, electronic device and storage medium to at least solve the problems of the existing entity weight scoring method, such as the cumbersome data processing process, time-consuming manual processing, the lack of scoring standards, and the lack of statistics on evaluation data of entity attributes, which results in different and one-sided scoring.

[0005] In the first aspect, an embodiment of the present application provides an entity weight scoring method, including: a standard library establishment step, establishing a standard library, setting standard entity attributes, standard emotional states, and weight values ​​of the standard emotional states in the standard library; a data acquisition step, acquiring the original evaluation data of a product on the e-commerce platform; a data preprocessing step, filtering the original evaluation data for redundant information and performing word segmentation processing; an entity processing step, performing entity-level processing on the preprocessed original evaluation data, and establishing an entity-emotion correspondence table based on the entity-level processing results; a scoring step, calculating the score of the product based on the entity-emotion correspondence table and the standard library.

[0006] Preferably, the entity processing step includes: an entity labeling step, labeling the entities and entity attributes in the original evaluation data, and obtaining the emotional state of the entity attributes; an attribute matching step, comparing the entity attributes with the standard entity attributes in the standard library, when the similarity reaches a threshold, determining that the entity attributes are similar, and recording the matched standard entity attributes; an emotion matching step, comparing the emotional state of the entity attributes with the standard emotional state in the standard library, when the similarity reaches a threshold, determining that the emotional state is similar, and recording the matched standard emotional state; a correspondence table establishment step, establishing the entity, the standard entity attributes corresponding to the entity, and the standard emotional state into an entity emotion correspondence table.

[0007] Preferably, the entity attribute labeling is implemented through a LSTM-CRF model.

[0008] Preferably, the similarity matching is achieved by cosine similarity matching.

[0009] In the second aspect, an embodiment of the present application provides an entity weight scoring system, which is applicable to the above-mentioned entity weight scoring method, including: a standard library establishment unit, including a standard library, in which standard entity attributes, standard emotional states, and weight values ​​of the standard emotional states are set; a data acquisition unit, which acquires the original evaluation data of a product on the e-commerce platform; a data preprocessing unit, which filters redundant information and performs word segmentation on the original evaluation data; an entity processing unit, which performs entity-level processing on the preprocessed original evaluation data, and establishes an entity emotion correspondence table based on the entity-level processing results; a scoring unit, which calculates the score of the product based on the entity emotion correspondence table and the standard library.

[0010] In some embodiments, the entity processing unit includes: an entity labeling module, which labels the entities and entity attributes in the original evaluation data and obtains the emotional state of the entity attributes; an attribute matching module, which compares the entity attributes with the standard entity attributes in the standard library, and when the similarity reaches a threshold, it is determined that the entity attributes are similar, and the matched standard entity attributes are recorded; an emotion matching module, which compares the emotional state of the entity attributes with the standard emotional state in the standard library, and when the similarity reaches a threshold, it is determined that the emotional state is similar, and the matched standard emotional state is recorded; a correspondence table establishment module, which establishes the entity, the standard entity attributes corresponding to the entity, and the standard emotional state into an entity emotion correspondence table.

[0011] In some of the embodiments, the entity attribute labeling is implemented through a LSTM-CRF model.

[0012] In some of the embodiments, the similarity matching is achieved by cosine similarity matching.

[0013] In a third aspect, an embodiment of the present application provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, an entity weight scoring method as described in the first aspect above is implemented.

[0014] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements an entity weight scoring method as described in the first aspect above.

[0015] Compared with the related art, the entity weight scoring method provided in the embodiment of the present application solves the tediousness and objectivity of traditional manual labeling by using the LSTM-CRF model to label the entities in the evaluation data of the product, thereby greatly saving labor costs. At the same time, the model results are matched with similarity according to the existing standard library to standardize the attributes, and finally a weight scoring mechanism is used to generate the final score of the entity and its attributes, so that the entity score is based on the importance of its attributes as an indicator affecting its score, reducing the possibility of obtaining one-sided product review attitudes, and finally feeding back the scoring results to the brand advertisers of the product to provide them with guidance on improving entity performance and advertising investment. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0017] Figure 1 is a flow chart of the entity weight scoring method of the present invention;

[0018] Figure 2 for Figure 1 The step-by-step flow chart of step S4;

[0019] Figure 3 It is a framework diagram of the entity weight scoring system of the present invention;

[0020] Figure 4 A framework diagram of an electronic device of the present invention;

[0021] In the above picture:

[0022] 1. Standard library establishment unit; 2. Data acquisition unit; 3. Data preprocessing unit; 4. Entity processing unit; 5. Scoring unit; 41. Entity annotation module; 42. Attribute matching module; 43. Emotion matching module; 60. Bus; 61. Processor; 62. Memory; 63. Communication interface. DETAILED DESCRIPTION

[0023] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is described and illustrated below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. Based on the embodiments provided in the present application, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present application.

[0024] Obviously, the drawings described below are only some examples or embodiments of the present application. For ordinary technicians in this field, the present application can also be applied to other similar scenarios based on these drawings without creative work. In addition, it can also be understood that although the efforts made in this development process may be complicated and lengthy, for ordinary technicians in this field related to the content disclosed in this application, some changes in design, manufacturing or production based on the technical content disclosed in this application are just conventional technical means, and should not be understood as insufficient content disclosed in this application.

[0025] Reference to "embodiments" in this application means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those of ordinary skill in the art that the embodiments described in this application may be combined with other embodiments without conflict.

[0026] Unless otherwise defined, the technical terms or scientific terms involved in this application should be understood by people with ordinary skills in the technical field to which this application belongs. The words "one", "a", "a kind of", "the" and the like involved in this application do not indicate a quantitative limitation and may represent the singular or plural. The terms "include", "comprise", "have" and any of their variations involved in this application are intended to cover non-exclusive inclusions; for example, a process, method, system, product or device comprising a series of steps or modules (units) is not limited to the listed steps or units, but may also include steps or units that are not listed, or may also include other steps or units inherent to these processes, methods, products or devices.

[0027] Natural language processing is an important direction in the fields of computer science and artificial intelligence. It studies various theories and methods that can achieve effective communication between people and computers using natural language. Natural language processing is a science that integrates linguistics, computer science, and mathematics. Therefore, research in this field will involve natural language, that is, the language people use in daily life, so it is closely related to the study of linguistics, but there are important differences. Natural language processing is not a general study of natural language, but the development of computer systems that can effectively realize natural language communication, especially the software systems therein. Therefore, it is a part of computer science. Natural language processing (NLP) is a field of computer science, artificial intelligence, and linguistics that focuses on the interaction between computers and human (natural) language.

[0028] The conditional random field (CRF) was proposed by Lafferty et al. in 2001. It combines the characteristics of the maximum entropy model and the hidden Markov model. It is an undirected graph model. In recent years, it has achieved good results in sequence labeling tasks such as word segmentation, part-of-speech tagging, and named entity recognition. The conditional random field is a typical discriminant model. Its joint probability can be written as a combination of several potential functions, among which the most commonly used is the linear chain conditional random field. If x = (x1, x2, ... xn) represents the observed input data sequence, and y = (y1, y2, ... yn) represents a state sequence, given an input sequence, the linear chain CRF model defines the joint conditional probability of the state sequence as:

[0029] p(y|x)=exp{}(2-14);

[0030] Z(x) = {}(2-15);

[0031] Among them: Z is the probability normalization factor conditioned on the observation sequence x; fj(yi-1, yi, x, i) is an arbitrary characteristic function; is the weight of each characteristic function.

[0032] Long Short-Term Memory (LSTM) is a time-recurrent neural network designed to solve the long-term dependency problem of general RNN (recurrent neural network). All RNNs have a chain form of repeated neural network modules. In the standard RNN, this repeated structural module has only a very simple structure, such as a tanh layer. In recent years, with the rise of artificial intelligence, the LSTM neural network model has improved the shortcomings of the long-term dependency limitations of the ordinary recurrent neural network model, and has been widely used in various tasks of natural language processing and achieved good results.

[0033] A web crawler is a program that automatically extracts web pages. It downloads web pages from the World Wide Web for search engines and is an important component of search engines. Traditional crawlers start with the URLs of one or several initial web pages, obtain the URLs on the initial web pages, and in the process of crawling web pages, continuously extract new URLs from the current page and put them into the queue until a certain stop condition of the system is met. The workflow of a focused crawler is more complicated. It needs to filter out links that are not related to the topic according to a certain web page analysis algorithm, retain useful links and put them into the queue of URLs waiting to be crawled. Then, it will select the next web page URL to be crawled from the queue according to a certain search strategy, and repeat the above process until a certain condition of the system is met. In addition, all web pages crawled by the crawler will be stored by the system, analyzed, filtered, and indexed for subsequent query and retrieval; for a focused crawler, the analysis results obtained in this process may also provide feedback and guidance for the subsequent crawling process.

[0034] Cosine similarity measures the similarity between two vectors by measuring the cosine of the angle between them. The cosine of an angle of 0 degrees is 1, while the cosine of any other angle is not greater than 1; and its minimum value is -1. Therefore, the cosine of the angle between two vectors determines whether the two vectors point in approximately the same direction. When two vectors have the same direction, the cosine similarity value is 1; when the angle between the two vectors is 90°, the cosine similarity value is 0; when the two vectors point in completely opposite directions, the cosine similarity value is -1. This result is independent of the length of the vectors, only related to the direction of the vectors. Cosine similarity is usually used in positive space, so the value given is between -1 and 1. Note that the upper and lower bounds are applicable to vector spaces of any dimension, and cosine similarity is most commonly used in high-dimensional positive spaces. For example, in information retrieval, each term is assigned a different dimension, and a dimension is represented by a vector whose values ​​in each dimension correspond to the frequency of the term in the document. Cosine similarity can therefore give the similarity of two documents in terms of their topics. In addition, it is often used for document comparison in text mining. In addition, in the field of data mining, it is used to measure the cohesion within a cluster.

[0035] Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings:

[0036] Figure 1 For the flow chart of the entity weight scoring method of the present invention, please refer to Figure 1 The entity weight scoring method of the present invention comprises the following steps:

[0037] S1: First, a standard library is established, and standard entity attributes, standard emotional states, and weight values ​​of standard emotional states are independently set in the standard library according to actual conditions.

[0038] S2: Obtain original evaluation data of a product on the e-commerce platform; optionally, the original evaluation data of the product may be automatically obtained through crawler technology.

[0039] S3: The original evaluation data may be filtered for redundant information according to factors such as time, and the evaluation data may be segmented for subsequent entity labeling.

[0040] S4: performing entity-level processing on the pre-processed original evaluation data, and establishing an entity-emotion correspondence table according to the entity-level processing result; optionally, Figure 2 for Figure 1 For a step-by-step flowchart of step S4 in Figure 2 :

[0041] S41. Labeling entities and entity attributes in the original evaluation data, and obtaining the sentiment state of the entity attributes; optionally, an LSTM-CRF model may be used for automatic entity attribute labeling, that is, the LSTM-CRF model may be used when performing named entity recognition on entities;

[0042] S42. Compare the entity attributes with the standard entity attributes in the standard library. When the similarity reaches a threshold, determine that the entity attributes are similar, and record the matched standard entity attributes; optionally, matching can be performed here by a cosine similarity matching method;

[0043] S43. Compare the emotional state of the entity attribute with the standard emotional state in the standard library. When the similarity reaches a threshold, it is determined that the emotional state is similar, and the matched standard emotional state is recorded; optionally, the matching can be performed here by a cosine similarity matching method;

[0044] The above cosine similarity matching method has the following formula:

[0045]

[0046] Where A is the entity attribute and B is the standard entity attribute; or, A is the emotional state and B is the standard emotional state;

[0047] S44. Establish an entity-emotion correspondence table by including the entity, the standard entity attributes corresponding to the entity, and the standard emotional state.

[0048] S5: Calculate the score of the product according to the entity emotion correspondence table and the standard library; optionally, first calculate the entity attribute evaluation value, the formula is:

[0049] V(v1,v2,v3)=W{w1,w2,w3,…}*E{e attribute1 ,e attribute2 ,e attribute3 ,…}

[0050] Where W represents the weight value of the entity attribute, and E represents the weight value of the emotional state in the standard library. Since each entity attribute has different importance in the entity, a different weight value W is set according to its importance. The more important the entity attribute is, the larger the weight value W is. Where w1+w2+…+w n =1, when the emotional state corresponding to the entity attribute does not exist, a preset default emotional state weight value constant is selected;

[0051] Optionally, the final score is obtained based on the average of the evaluation values ​​of each entity attribute of the original evaluation data.

[0052] It should be noted that the steps shown in the above process or the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0053] The present application embodiment provides an entity weight scoring system, which is applicable to the above-mentioned entity weight scoring method. As used below, the terms "unit", "module", etc. can implement a combination of software and / or hardware for a predetermined function. Although the device described in the following embodiments is preferably implemented in software, the implementation of hardware or a combination of software and hardware is also possible and conceived.

[0054] Figure 3 For a framework diagram of the entity weight scoring system according to the present invention, see Figure 3 ,include:

[0055] The standard library establishment unit 1 includes a pre-established standard library, in which standard entity attributes, standard emotional states, and weight values ​​of standard emotional states are independently set according to actual conditions.

[0056] The data acquisition unit 2 acquires the original evaluation data of a product on the e-commerce platform; optionally, the original evaluation data of the above-mentioned product can be automatically acquired through crawler technology.

[0057] The data preprocessing unit 3 may filter redundant information from the original evaluation data according to factors such as time, and perform word segmentation on the evaluation data for subsequent entity labeling.

[0058] The entity processing unit 4 performs entity level processing on the pre-processed original evaluation data, and establishes an entity emotion correspondence table according to the entity level processing result; optionally, the entity processing unit 4 also includes:

[0059] The entity labeling module 41 labels the entities and entity attributes in the original evaluation data, and obtains the sentiment state of the entity attributes; optionally, the LSTM-CRF model can be used for automatic entity attribute labeling, that is, the LSTM-CRF model is used when performing named entity recognition on the entity;

[0060] The attribute matching module 42 compares the entity attributes with the standard entity attributes in the standard library, and when the similarity reaches a threshold, determines that the entity attributes are similar, and records the matched standard entity attributes; optionally, the matching can be performed here by a cosine similarity matching method;

[0061] The emotion matching module 43 compares the emotion state of the entity attribute with the standard emotion state in the standard library, and when the similarity reaches a threshold, determines that the emotion state is similar, and records the matched standard emotion state; optionally, the matching can be performed here by a cosine similarity matching method;

[0062] The above cosine similarity matching method has the following formula:

[0063]

[0064] Where A is the entity attribute and B is the standard entity attribute; or, A is the emotional state and B is the standard emotional state;

[0065] The corresponding table establishment module 44 establishes an entity emotion corresponding table by combining the entity, the standard entity attribute corresponding to the entity, and the standard emotion state. The scoring unit 5 calculates the score of the product according to the entity emotion corresponding table and the standard library; optionally, the entity attribute evaluation value is calculated first, and the formula is:

[0066] V(v1,v2,v3)=W{w1,w2,w3,…}*E{e attribute1 ,e attribute2 ,e attribute3 ,…}

[0067] Where W represents the weight value of the entity attribute, and E represents the weight value of the emotional state in the standard library. Since each entity attribute has different importance in the entity, a different weight value W is set according to its importance. The more important the entity attribute is, the larger the weight value W is. Where w1+w2+…+w n =1, when the emotional state corresponding to the entity attribute does not exist, a preset default emotional state weight value constant is selected;

[0068] Optionally, the final score is obtained based on the average of the evaluation values ​​of each entity attribute of the original evaluation data.

[0069] In addition, combined Figure 1 , Figure 2 The described entity weight scoring method can be implemented by an electronic device. Figure 4 It is a framework diagram of the electronic device of the present invention.

[0070] The electronic device may include a processor 61 and a memory 62 storing computer program instructions.

[0071] Specifically, the processor 61 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.

[0072] Among them, the memory 62 may include a large capacity memory for data or instructions. By way of example and not limitation, the memory 62 may include a hard disk drive (HDD), a floppy disk drive, a solid state drive (SSD), a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive, or a combination of two or more of these. Where appropriate, the memory 62 may include a removable or non-removable (or fixed) medium. Where appropriate, the memory 62 may be inside or outside the data processing device. In a specific embodiment, the memory 62 is a non-volatile memory. In a specific embodiment, the memory 62 includes a read-only memory (ROM) and a random access memory (RAM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (Programmable Read-Only Memory, PROM for short), an erasable PROM (Erasable ProgrammableRead-Only Memory, EPROM for short), an electrically erasable PROM (Electrically Erasable ProgrammableRead-Only Memory, EEPROM for short), an electrically alterable ROM (Electrically Alterable Read-Only Memory, EAROM for short) or a flash memory (FLASH) or a combination of two or more of these. Under appropriate circumstances, the RAM can be a static random access memory (SRAM) or a dynamic random access memory (DRAM), wherein the DRAM can be a fast page mode dynamic random access memory (FPMDRAM), an extended data output dynamic random access memory (EDODRAM), a synchronous dynamic random access memory (SDRAM), etc.

[0073] The memory 62 may be used to store or cache various data files required for processing and / or communication, as well as possible computer program instructions executed by the processor 61 .

[0074] The processor 61 implements any one of the entity weight scoring methods in the above embodiments by reading and executing computer program instructions stored in the memory 62 .

[0075] In some of the embodiments, the electronic device may further include a communication interface 63 and a bus 60. Figure 4 As shown, the processor 61, the memory 62, and the communication interface 63 are connected via a bus 60 and communicate with each other.

[0076] The communication port 63 can realize data communication with other components such as external devices, image / data acquisition devices, databases, external storage, and image / data processing workstations.

[0077] The bus 60 includes hardware, software or both, and couples the components of the electronic device to each other. The bus 60 includes but is not limited to at least one of the following: a data bus, an address bus, a control bus, an expansion bus, and a local bus. By way of example and not limitation, bus 60 may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an InfiniBand interconnect, a Low Pin Count (LPC) bus, a memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local Bus (VLB) bus, or other suitable buses or a combination of two or more of these. Where appropriate, bus 60 may include one or more buses. Although embodiments of the present application describe and illustrate a particular bus, the present application contemplates any suitable bus or interconnect.

[0078] The electronic device can execute an entity weight scoring method in an embodiment of the present application.

[0079] In addition, in combination with an entity weight scoring method in the above embodiment, the present application embodiment can provide a computer-readable storage medium for implementation. The computer-readable storage medium stores computer program instructions; when the computer program instructions are executed by a processor, any entity weight scoring method in the above embodiment is implemented.

[0080] The aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk and other media that can store program codes.

[0081] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0082] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. An entity weight scoring method, characterized in that: include: A standard library establishment step is to establish a standard library, and set standard entity attributes, standard emotional states, and weight values ​​of the standard emotional states in the standard library; Data acquisition step, obtaining the original evaluation data of a product on the e-commerce platform; A data preprocessing step, filtering the raw evaluation data for redundant information and performing word segmentation processing; An entity processing step is to perform entity level processing on the pre-processed original evaluation data, and to establish an entity emotion correspondence table according to the entity level processing result; A scoring step, calculating the score of the product according to the entity emotion correspondence table and the standard library; The entity processing step includes: An entity labeling step, labeling entities and entity attributes in the original evaluation data, and obtaining the emotional state of the entity attributes; An attribute matching step, comparing the entity attributes with the standard entity attributes in the standard library, and when the similarity reaches a threshold, determining that the entity attributes are similar, and recording the matched standard entity attributes; The emotion matching step is to compare the emotion state of the entity attribute with the standard emotion state in the standard library, and when the similarity reaches a threshold, it is determined that the emotion state is similar, and the matched standard emotion state is recorded; A corresponding table establishing step, establishing the entity, the standard entity attribute corresponding to the entity, and the standard emotional state into an entity-emotion corresponding table; Wherein, the entity attribute labeling is implemented through an LSTM-CRF model; Among them, first calculate the entity attribute evaluation value, the formula is: V(v1,v2,v3)=W{w1,w2,w3,…}*E{eattribute 1,eatt ribute2,eatt ribute3,…}, where W represents the entity attribute weight value, E represents the weight value of the emotional state in the standard library, and each entity attribute has different importance in the entity, so different weight values ​​W are set according to their importance. The more important the entity attribute is, the larger the weight value W is, where w1+w2+…+wn=1, when the emotional state corresponding to the entity attribute does not exist, a preset default emotional state weight value constant is selected; The final score is then obtained based on the average value of each entity attribute evaluation value of the original evaluation data.

2. The entity weight scoring method according to claim 1, characterized in that: The similarity matching is achieved through cosine similarity matching.

3. An entity weight scoring system, comprising: A standard library establishment unit includes a standard library, in which standard entity attributes, standard emotional states, and weight values ​​of the standard emotional states are set; A data acquisition unit, which acquires the original evaluation data of a product on the e-commerce platform; A data preprocessing unit, filtering redundant information from the original evaluation data and performing word segmentation processing; An entity processing unit performs entity level processing on the pre-processed original evaluation data, and establishes an entity emotion correspondence table according to the entity level processing result; A scoring unit, which calculates the score of the product according to the entity emotion correspondence table and the standard library; Wherein, the entity processing unit includes: An entity annotation module annotates entities and entity attributes in the original evaluation data and obtains the sentiment state of the entity attributes; An attribute matching module compares the entity attributes with the standard entity attributes in the standard library, and when the similarity reaches a threshold, determines that the entity attributes are similar, and records the matched standard entity attributes; An emotion matching module compares the emotion state of the entity attribute with the standard emotion state in the standard library, and when the similarity reaches a threshold, determines that the emotion states are similar, and records the matched standard emotion state; A correspondence table establishment module establishes the entity, the standard entity attribute corresponding to the entity, and the standard emotion state into an entity emotion correspondence table; Wherein, the entity attribute labeling is implemented through an LSTM-CRF model; Among them, first calculate the entity attribute evaluation value, the formula is: V(v1,v2,v3)=W{w1,w2,w3,…}*E{eattribute 1,eatt ribute2,eatt ribute3,…}, where W represents the entity attribute weight value, E represents the weight value of the emotional state in the standard library, and each entity attribute has different importance in the entity, so different weight values ​​W are set according to their importance. The more important the entity attribute is, the larger the weight value W is, where w1+w2+…+wn=1, when the emotional state corresponding to the entity attribute does not exist, a preset default emotional state weight value constant is selected; The final score is then obtained based on the average value of each entity attribute evaluation value of the original evaluation data.

4. The entity weight scoring system according to claim 3, characterized in that: The similarity matching is achieved through cosine similarity matching.

5. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the entity weight scoring method according to any one of claims 1 to 2 is implemented.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the entity weight scoring method according to any one of claims 1 to 2 is implemented.

Images