Evaluation method, computing device and computer readable storage medium

Abstract

The invention discloses an evaluation method, computing equipment and a computer readable storage medium, and the method comprises the steps: carrying out the semantic and sentiment analysis of a target file of a target object, and determining a language evaluation value of the target file; determining a decision matrix according to the language evaluation value and the data set of the target object; obtaining an ideal evaluation value, and determining a preference distance measure according to the ideal evaluation value and the decision matrix; fitting calculation is carried out based on the preference distance measure, and an evaluation result of the target object is determined. Therefore, the target file is subjected to semantic and sentiment analysis, fitting calculation is performed by determining the preference distance measure, and the corresponding evaluation result is determined, so that objective analysis of the target file is realized, and the accuracy of the evaluation result is improved.

Description

Technical Field

[0001] The present application relates to the technical field of data processing, and in particular to an evaluation method, a computing device, and a computer-readable storage medium. Background Art

[0002] When enterprises or individuals currently obtain evaluation results of target objects such as suppliers, they generally assign weights to various indicators of suppliers based on the analytic hierarchy process (AHP), or develop an evaluation index system and scoring system based on the supplier performance scorecard to determine the evaluation results of each supplier. Among them, supplier evaluation refers to the use of an index evaluation system to evaluate the supplier's supply quality service level, supply price, punctuality, creditworthiness, etc., laying the foundation for supplier selection.

[0003] However, in reality, since evaluators mostly use vague terms with inaccurate information to evaluate the target objects, and they only conduct a single evaluation at the end, the final evaluation will be subjective and the results will not be accurate enough. Summary of the invention

[0004] The purpose of this application is to provide an evaluation method, a computing device, and a computer-readable storage medium, which achieve objective analysis of the target file and improve the accuracy of the evaluation results.

[0005] To achieve the above objectives:

[0006] In a first aspect, the present application provides an evaluation method, comprising:

[0007] Performing semantic and sentiment analysis on the target file of the target object to determine the language evaluation value of the target file;

[0008] Determining a decision matrix based on the language evaluation value and the data set of the target object;

[0009] Obtaining an ideal evaluation value, and determining a preference distance measure according to the ideal evaluation value and the decision matrix;

[0010] A fitting calculation is performed based on the preference distance measure to determine an evaluation result of the target object.

[0011] In one embodiment, the performing semantic and sentiment analysis on the target file of the target object to determine the language evaluation value of the target file includes:

[0012] Determining a feature vector of the target file;

[0013] Inputting the feature vector into a preset analysis model for semantic and sentiment analysis to obtain an evaluation vector;

[0014] According to the evaluation vector, based on preset evaluation indicators and language evaluation scale, the language evaluation value of the target file is determined; the evaluation indicator is used to characterize the dimension of evaluation; the language evaluation scale is used to characterize the grade standard of evaluation.

[0015] In one implementation, determining the feature vector of the target file includes:

[0016] Performing text preprocessing on the target file;

[0017] Based on word embedding technology, the preprocessed target file is converted into a feature vector.

[0018] In one embodiment, determining a decision matrix based on the language evaluation value and the target object data set includes:

[0019] A decision matrix is ​​determined according to the language evaluation value and the data set of the target object based on a conversion function.

[0020] In one embodiment, obtaining the ideal evaluation value includes:

[0021] Acquiring evaluation data, and determining a positive ideal point and a negative ideal point of the language evaluation value based on the evaluation data;

[0022] If the linguistic values ​​of the positive ideal point and the negative ideal point are discontinuous, then the positive ideal point is expanded into a continuous positive ideal data set, and the negative ideal point is expanded into a continuous negative ideal data set;

[0023] The positive ideal data set and the negative ideal data set are determined as ideal evaluation values.

[0024] In one embodiment, the ideal evaluation value includes a positive ideal point and a negative ideal point; the preference distance measure includes a first preference measure and a second preference measure;

[0025] Determining a preference distance measure according to the ideal evaluation value and the decision matrix includes:

[0026] Determining a first preference measure of the positive ideal point relative to the decision matrix by a distance measure operator; and

[0027] A second preference measure of the negative ideal point relative to the decision matrix is ​​determined.

[0028] In one embodiment, the method for determining the distance measurement operator includes:

[0029] Calculate the deviation of the ideal evaluation value relative to each element in the decision matrix according to the preset weights;

[0030] The deviation values ​​are summed up based on preset weights to determine a distance measurement operator.

[0031] In one implementation, performing the fitting calculation based on the preference distance measure to determine the evaluation result of the target object includes:

[0032] An evaluation result of the target object is determined according to the second preference measure and the sum of the first preference measure and the second preference measure.

[0033] In a second aspect, an embodiment of the present application provides a computing device, specifically comprising:

[0034] processor;

[0035] a memory for storing instructions executable by the processor;

[0036] The processor is configured to execute the instructions to perform the evaluation method as described in the first aspect.

[0037] In a third aspect, an embodiment of the present application provides a computer-readable storage medium, in which a computer program is stored. When the instructions in the computer-readable storage medium are executed by a processor of a computing device, the computing device is enabled to implement the evaluation method described in the first aspect.

[0038] The evaluation method, computing device, and computer-readable storage medium provided by the embodiment of the present application include: performing semantic and sentiment analysis on the target file of the target object to determine the language evaluation value of the target file; determining a decision matrix based on the language evaluation value and the data set of the target object; obtaining an ideal evaluation value, and determining a preference distance measure based on the ideal evaluation value and the decision matrix; performing a fitting calculation based on the preference distance measure to determine the evaluation result of the target object. In this way, by performing semantic and sentiment analysis on the target file, and performing a fitting calculation by determining the preference distance measure to determine the corresponding evaluation result, an objective analysis of the target file is achieved, and the accuracy of the evaluation result is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 A schematic diagram of a flow chart of an evaluation method provided by an embodiment of the present invention;

[0040] Figure 2 A schematic diagram of the structure of a computing device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0041] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0042] It should be noted that, in this article, the terms "include", "comprises" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element. In addition, components, features, and elements with the same name in different embodiments of the present application may have the same meaning or different meanings, and their specific meanings need to be determined by their explanation in the specific embodiment or further combined with the context of the specific embodiment.

[0043] It should be understood that, although the terms first, second, third, etc. may be used to describe various information in this article, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of this article, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein can be interpreted as "at the time of..." or "when..." or "in response to determination". Furthermore, as used in this article, the singular forms "one", "one" and "the" are intended to also include plural forms, unless there is an opposite indication in the context. It should be further understood that the terms "comprising", "including" indicate that there are described features, steps, operations, elements, components, projects, kinds, and / or groups, but do not exclude the existence, occurrence or addition of one or more other features, steps, operations, elements, components, projects, kinds, and / or groups. The terms "or" and "and / or" used herein are interpreted as inclusive, or mean any one or any combination. Thus, “A, B, or C” or “A, B and / or C” means “any of the following: A; B; C; A and B; A and C; B and C; A, B, and C.” An exception to this definition will occur only when a combination of elements, functions, steps, or operations are inherently mutually exclusive in some manner.

[0044] It should be understood that, although the various steps in the flowchart in the embodiment of the present application are displayed in sequence according to the indication of the arrows, these steps are not necessarily performed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and it can be performed in other orders. Moreover, at least a portion of the steps in the figure may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and their execution order is not necessarily performed in sequence, but can be performed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.

[0045] It should be noted that in this article, step codes such as S101, S102, etc. are used for the purpose of expressing the corresponding content more clearly and concisely, and do not constitute a substantial limitation on the sequence. When implementing the step, those skilled in the art may execute S102 first and then S101, etc., but these should all be within the scope of protection of this application.

[0046] It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0047] In the subsequent description, the suffixes such as "module", "component" or "unit" used to represent elements are only used to facilitate the description of the present application and have no specific meaning. Therefore, "module", "component" or "unit" can be used in a mixed manner.

[0048] See also Figure 1 The embodiment of the present application provides an evaluation method, which can be implemented in software and / or hardware. In this embodiment, the evaluation method is applied to a computing device as an example. The evaluation method provided in this embodiment includes:

[0049] Step S101: Perform semantic and sentiment analysis on the target file of the target object to determine the language evaluation value of the target file.

[0050] The target file includes daily communication data between the evaluator and the target object, including recording content and text content. Here, daily communication data includes but is not limited to communication data generated in various business scenarios such as sourcing, daily communication, and business communication. Optionally, the daily communication data generated within a preset time interval can be obtained and determined as the target file.

[0051] For example, take the target object as a supplier. Since the supplier evaluation frequency is monthly, quarterly or even annual, the chat records of daily communication can be obtained based on a preset time interval, such as one month, one quarter, one year, etc., and the chat records can be determined as the target file.

[0052] In one implementation, the target file of the target object is subjected to semantic and sentiment analysis to determine the language evaluation value of the target file, including:

[0053] Determine the feature vector of the target file;

[0054] Input the feature vector into the preset analysis model for semantic and sentiment analysis to obtain the evaluation vector;

[0055] According to the evaluation vector, based on the preset evaluation index and language evaluation scale, the language evaluation value of the target file is determined; the evaluation index is used to characterize the evaluation dimension; the language evaluation scale is used to characterize the evaluation grade standard.

[0056] Among them, semantic analysis refers to understanding the deep meaning of a sentence or text based on a preset feature vector, including entity recognition (such as identifying entities in the text, including names of people, places, etc.), relationship extraction (such as extracting the relationship between entities), sentiment analysis (such as judging the sentiment tendency of the text), etc.

[0057] Sentiment analysis refers to the recognition of emotions in the language scenarios and language information before and after key nodes of the target object's evaluation, and the mining of speech frequency characteristics (such as pitch shape, average pitch, pitch contour, tail drop and range), time-related characteristics (such as speaking speed, stress frequency), sound quality parameters and energy descriptors (such as breathing sounds, brightness, loudness, pause discontinuity, pitch discontinuity) in this scenario.

[0058] Optionally, the preset analysis model can be trained based on preset test data or historical evaluation data to determine the parameter data for evaluation by the analysis model. In this way, the feature vector of the target file is input into the analysis model to perform semantic analysis and sentiment analysis on the feature vector based on the analysis model, and output the evaluation vector as the analysis result. In this way, by performing sentiment analysis, it helps to understand the evaluator's positive or negative feelings towards the target object. At the same time, by performing semantic analysis, it helps to determine the specific needs or standards on which the evaluator makes the evaluation, thereby providing deeper insights.

[0059] Among them, the evaluation index is used to characterize the evaluation dimension, that is, the evaluation index is determined by the evaluation index. Here, the corresponding dimensions for different industries may be different. For example, in the evaluation of suppliers, the evaluation indicators can be set to include quality dimension, cost dimension, service dimension, payment dimension and cooperation dimension, etc. At the same time, the secondary dimensions can be further refined based on these dimensions for analysis. The specific evaluation indicators are shown in Table 1.

[0060] Among them, the language evaluation scale is used to characterize the evaluation grade standard, that is, to construct a multi-granularity fuzzy language evaluation scale based on the preset evaluation indicators. This scale will contain multiple grade standards, each grade standard corresponds to a fuzzy set, which is used to describe evaluations of different strengths. For example, five levels of "very bad", "bad", "average", "good" and "very good" can be set, and each grade standard has its boundaries and coverage defined by a membership function.

[0061] Table 1

[0062]

[0063]

[0064] Optionally, the corresponding evaluation dimension is determined in the evaluation index according to the obtained evaluation vector, and the corresponding grade standard is determined in the language evaluation scale to generate an evaluation value of a certain dimension. Here, the evaluation value is generally represented by a set. In this way, after obtaining the evaluation values ​​of multiple dimensions, the final language evaluation value is calculated by weighted average based on the preset weight values ​​corresponding to different evaluation dimensions. Here, when determining the evaluation index, the keywords and key information of these evaluations can be automatically extracted through text mining and semantic analysis to form a multi-dimensional comprehensive evaluation.

[0065] For example, taking the evaluation index as shown in Table 1 above, and the language evaluation scale represented by the seven-granularity fuzzy language evaluation scale as an example, the language evaluation scale is determined as: S = {S -3 = Extreme, S -2 = difference, S -1 = Poor, S0 = Average, S1 = Good, S2 = Good, S3 = Excellent}. In the process of communication between the buyer and supplier A, it is found that in terms of on-time delivery (i.e. the 12th evaluation indicator in the evaluation indicators), the performance of supplier A is identified as "average" and "good", then its evaluation value is S 12 ={S0, S1, S2}; It is found that in terms of cooperation attitude (i.e., the 15th evaluation index in the evaluation index), the performance of supplier A is at most "average", so its evaluation value is S 15 ={S -3 , S -2 , S -1 , S0}.

[0066] In one embodiment, determining a feature vector of a target file includes:

[0067] Perform text preprocessing on the target file;

[0068] Based on word embedding technology, the preprocessed target file is converted into a feature vector.

[0069] Among them, text preprocessing includes text conversion processing, data cleaning, word segmentation, part-of-speech tagging, etc. Specifically, it is expressed as follows: First, when the target file includes audio content, the audio content is subjected to text recognition to convert the target file into text language. Secondly, data cleaning steps are performed, including removing HTML tags, removing special characters, removing stop words, etc., to reduce noise and improve the efficiency of subsequent analysis. Among them, special characters can be non-text elements, such as emoticons, punctuation marks, etc. Then, considering that there are no obvious lexical boundaries in the text content, word segmentation processing is performed accordingly, that is, the continuous text is divided into independent lexical units, while considering the ambiguity and contextual information of the vocabulary to improve the accuracy of word segmentation. Finally, part-of-speech tagging is performed to determine the part of speech of each word to improve the accuracy of word segmentation.

[0070] Optionally, the target file obtained after preprocessing is converted based on word embedding technology, where word embedding is the process of converting vocabulary text into a vector representation that can be embedded in a computer. Common word embedding technologies include Word2Vec, GloVe, etc.

[0071] Step S102: Determine a decision matrix according to the language evaluation value and the data set of the target object.

[0072] Optionally, each element of the decision matrix is ​​a quantitative representation of the linguistic evaluation value of the target object on the corresponding attribute. When constructing the decision matrix, it is necessary to ensure that all target objects are evaluated for the same set of attributes to ensure the consistency and comparability of the matrix.

[0073] In one embodiment, a decision matrix is ​​determined based on the language evaluation value and the target object data set, including:

[0074] According to the language evaluation value and the data set of the target object, a decision matrix is ​​determined based on the conversion function.

[0075] Optionally, according to the language evaluation value and the target object data set, based on the conversion function, the decision matrix can be specifically expressed as: Set C = {C1, C2, ..., C n} is the attribute value of the language evaluation value, where n represents the n dimensions of the language evaluation value. Set ω=(ω(c1),ω(c2),...,ω(c n )) is the preset weight value corresponding to each of the above attribute values, where ω(c j )∈[0,1](j=1,2,...,n) and Set X = {x1, x2, ..., x m} is the target object. In this way, based on the decision maker's language expression of ∫∫(x i ,cj ) gives the target object x i (i=1,2,...,m) in attribute value C j (j=1,2,...,n) language evaluation value, and according to the conversion function ∫∫(x i ,c j ) is converted into the target fuzzy language term set In this way, multiple target fuzzy language term sets are determined Composition of the HFLTS decision matrix

[0076] Step S103: Obtain an ideal evaluation value, and determine a preference distance measure according to the ideal evaluation value and the decision matrix.

[0077] The ideal evaluation value represents the best or worst performance that can be achieved on a specific attribute. By comparing with the ideal evaluation value, the relative merits of the target fuzzy language term set of the target object can be measured.

[0078] In one embodiment, obtaining an ideal evaluation value includes:

[0079] Acquire evaluation data, and determine a positive ideal point and a negative ideal point of a language evaluation value based on the evaluation data;

[0080] If the linguistic values ​​of the positive ideal point and the negative ideal point are discontinuous, the positive ideal point is expanded into a continuous positive ideal data set, and the negative ideal point is expanded into a continuous negative ideal data set;

[0081] A positive ideal data set and a negative ideal data set are determined as ideal evaluation values.

[0082] Among them, a positive ideal point refers to a virtual target object that can achieve the best performance on all attributes. In the context of HFLTS, a positive ideal point usually consists of the best language terms on each attribute. A negative ideal point refers to a virtual target object that can achieve the worst performance on all attributes. In the context of HFLTS, a negative ideal point usually consists of the worst language terms on each attribute.

[0083] Optionally, collect raw data related to the decision, which may come from questionnaires, experimental observations, historical records, etc. Convert the raw data into a hesitant fuzzy language term set (HFLTS), and determine the positive ideal point and negative ideal point in the hesitant fuzzy language term set. Here, the positive ideal point H s + and negative ideal point H s - It can be specifically expressed as:

[0084]

[0085] In one embodiment, if and If the language value in is not continuous, then expand it into a continuous language data set and It is understandable that when the language values ​​in the positive ideal point and the negative ideal point are discontinuous, they need to be expanded into a continuous language data set, which can be achieved by adding new language values, which should be able to fill the blank areas between the original language values.

[0086] It can be understood that if the language values ​​of the positive ideal point and the negative ideal point are continuous, the positive ideal point and the negative ideal point are determined as ideal evaluation values; if the language values ​​of the positive ideal point and the negative ideal point are discontinuous, the expanded positive ideal data set and the negative ideal data set are determined as ideal evaluation values.

[0087] In one embodiment, the ideal evaluation value includes a positive ideal point and a negative ideal point; the preference distance measure includes a first preference measure and a second preference measure;

[0088] According to the ideal evaluation value and decision matrix, the preference distance measure is determined, including:

[0089] Determine the first preference measure of the positive ideal point relative to the decision matrix by a distance measure operator; and

[0090] Determine the second preference measure of the negative ideal point with respect to the decision matrix.

[0091] Optionally, use the distance measure operator to calculate each target object x separately i (i=1,2,...,m) The evaluation value of each attribute and the HFLTS positive ideal point The first preference measure between (also called the positive weighted mixed hesitant preference distance measure) and negative ideal point The second preference measure between (which can be called the positive weighted mixed hesitant preference distance measure) It can be calculated as:

[0092]

[0093] Among them, i=1, 2,…, m.

[0094] Among them, the method for determining the distance measurement operator includes:

[0095] According to the preset weights, calculate the deviation value of the ideal evaluation value relative to each element in the decision matrix;

[0096] The deviation values ​​are summed up based on preset weights to determine the distance measurement operator.

[0097] Specifically, the calculation method for determining the distance measurement operator can be specifically expressed by the formula:

[0098]

[0099] in, δ l is the rate of change of the subscript; γ and η are preset weights, 0≤γ+η≤1 and γ+η=1; ω i for The weight value, ω=(ω1,ω2,...,ω n ),ω i ≥0 and

[0100] Among them, ω i It can be a preset weight value or determined based on the following model (LP), which can be expressed by the formula:

[0101]

[0102] Among them, R j The calculation method of and R is expressed as:

[0103]

[0104] Step S104: Perform fitting calculation based on the preference distance measurement to determine the evaluation result of the target object.

[0105] In one embodiment, performing a fit calculation based on the preference distance measure to determine the evaluation result of the target object includes:

[0106] An evaluation result of the target object is determined according to the second preference measure, the sum of the first preference measure and the second preference measure.

[0107] Optionally, construct the target object x i (i=1,2,...,m) closeness, that is, according to the idea of ​​TOPSIS method, when the first preference measure The smaller the second preference measure The larger the value, the higher the evaluation result and the better the solution. Here, the calculation formula of closeness is expressed as:

[0108]

[0109] According to the closeness D i (i=1,2,...,m) for solution X i (i=1,2,...,m) are sorted. It can be understood that D i The larger the i The better.

[0110] In one embodiment, after the evaluation results of the target object are calculated and determined, the evaluation results of the target object are fed back to the supplier, so that the supplier can immediately understand the position and performance of their own products in the supply chain, and help the supplier identify the advantages and disadvantages of their own products, so as to make targeted and timely improvements.

[0111] In summary, in the evaluation method provided in the above embodiment, by performing semantic and sentiment analysis on the target file and determining the preference distance measurement for fitting calculation, the corresponding evaluation result is determined, thereby achieving objective analysis of the target file, helping to reduce the impact of human bias, providing more objective and fair evaluation results, and improving the accuracy of the evaluation results, so as to further help enterprises or institutions to make decisions based on more comprehensive and neutral data for analysis.

[0112] Based on the same inventive concept as the above embodiments, an embodiment of the present invention provides a computing device, such as Figure 2 As shown, the computing device includes: a processor 210 and a memory 211 storing a computer program; wherein, Figure 2 The processor 210 shown in the figure is not used to indicate that the number of the processor 210 is one, but is only used to indicate the positional relationship of the processor 210 relative to other devices. In actual applications, the number of the processor 210 may be one or more; similarly, Figure 2 The memory 211 shown in the figure has the same meaning, that is, it is only used to refer to the position relationship of the memory 211 relative to other devices. In practical applications, the number of memories 211 can be one or more. When the processor 210 runs the computer program, the above evaluation method is implemented.

[0113] The computing device may also include: at least one network interface 212. The various components in the computing device are coupled together via a bus system 213. It is understood that the bus system 213 is used to achieve connection and communication between these components. In addition to the data bus, the bus system 213 also includes a power bus, a control bus, and a status signal bus. However, for the sake of clarity, the bus system 213 is not used in the example above. Figure 2 Various buses are labeled as bus system 213.

[0114] The memory 211 may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. The non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a magnetic random access memory (FRAM), a flash memory, a magnetic surface memory, an optical disk, or a compact disc read-only memory (CD-ROM); the magnetic surface memory may be a disk memory or a tape memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static random access memory (SRAM), synchronous static random access memory (SSRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM, SyncLink Dynamic Random Access Memory), and direct RAM bus random access memory (DRRAM, Direct Rambus Random Access Memory).The memory 211 described in the embodiments of the present invention is intended to include but is not limited to these and any other suitable types of memories.

[0115] The memory 211 in the embodiment of the present invention is used to store various types of data to support the operation of the computing device. Examples of these data include: any computer program used to operate on the computing device, such as an operating system and an application; contact data; phone book data; messages; pictures; videos, etc. Among them, the operating system includes various system programs, such as a framework layer, a core library layer, a driver layer, etc., which are used to implement various basic services and process hardware-based tasks. The application program may include various applications, such as a media player, a browser, etc., which are used to implement various application services. Here, the program that implements the method of the embodiment of the present invention may be included in the application program.

[0116] Based on the same inventive concept as the above-mentioned embodiment, this embodiment further provides a computer-readable storage medium, in which a computer program is stored. The computer-readable storage medium may be a ferromagnetic random access memory (FRAM), a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a flash memory, a magnetic surface memory, an optical disc, or a compact disc read-only memory (CD-ROM); it may also be various devices including one or any combination of the above memories, such as a mobile phone, a computer, a tablet device, a personal digital assistant, etc. When the computer program stored in the computer-readable storage medium is executed by the processor, the evaluation method applied to the above-mentioned computing device is implemented. For the specific steps implemented when the computer program is executed by the processor, please refer to Figure 1 The description of the illustrated embodiment will not be repeated here.

[0117] 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.

[0118] In this document, the terms "comprises," "comprising," or any other variations thereof, are intended to cover a non-exclusive inclusion of elements other than those listed and may also include additional elements not expressly listed.

[0119] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. An evaluation method, characterized in that: include: Performing semantic and sentiment analysis on the target file of the target object to determine the language evaluation value of the target file; Determining a decision matrix according to the language evaluation value and the data set of the target object; Obtaining an ideal evaluation value, and determining a preference distance measure according to the ideal evaluation value and the decision matrix; A fitting calculation is performed based on the preference distance measure to determine an evaluation result of the target object.

2. The method according to claim 1, characterized in that The step of performing semantic and sentiment analysis on the target file of the target object to determine the language evaluation value of the target file includes: Determining a feature vector of the target file; Inputting the feature vector into a preset analysis model for semantic and sentiment analysis to obtain an evaluation vector; According to the evaluation vector, based on preset evaluation indicators and language evaluation scale, the language evaluation value of the target file is determined; the evaluation indicator is used to characterize the dimension of evaluation; the language evaluation scale is used to characterize the grade standard of evaluation.

3. The method according to claim 2, characterized in that The step of determining the feature vector of the target file comprises: Performing text preprocessing on the target file; Based on word embedding technology, the preprocessed target file is converted into a feature vector.

4. The method according to claim 1, characterized in that: The step of determining a decision matrix according to the language evaluation value and the target object data set includes: A decision matrix is ​​determined according to the language evaluation value and the data set of the target object based on a conversion function.

5. The method according to claim 1, characterized in that The obtaining of the ideal evaluation value comprises: Acquiring evaluation data, and determining a positive ideal point and a negative ideal point of the language evaluation value based on the evaluation data; If the linguistic values ​​of the positive ideal point and the negative ideal point are discontinuous, then the positive ideal point is expanded into a continuous positive ideal data set, and the negative ideal point is expanded into a continuous negative ideal data set; The positive ideal data set and the negative ideal data set are determined as ideal evaluation values.

6. The method according to claim 1, characterized in that The ideal evaluation value includes a positive ideal point and a negative ideal point; the preference distance measure includes a first preference measure and a second preference measure; Determining a preference distance measure according to the ideal evaluation value and the decision matrix includes: Determining a first preference measure of the positive ideal point relative to the decision matrix by a distance measure operator; as well as, A second preference measure of the negative ideal point relative to the decision matrix is ​​determined.

7. The method according to claim 6, characterized in that The method for determining the distance measurement operator includes: Calculate the deviation of the ideal evaluation value relative to each element in the decision matrix according to the preset weights; The deviation values ​​are summed up based on preset weights to determine a distance measurement operator.

8. The method according to claim 6, characterized in that The performing the fitting calculation based on the preference distance measure to determine the evaluation result of the target object includes: An evaluation result of the target object is determined according to the second preference measure and the sum of the first preference measure and the second preference measure.

9. A computing device, characterized in that include: A processor and a memory for storing executable instructions; wherein the processor is configured to execute the instructions to implement the evaluation method according to any one of claims 1-8.

10. A computer-readable storage medium, characterized in that: When the instructions in the computer-readable storage medium are executed by a processor, the evaluation method according to any one of claims 1 to 8 is implemented.

Images