Service evaluation text generation method and device, computer equipment, readable storage medium and program product

Abstract

The invention relates to a service evaluation text generation method and device, computer equipment, a computer readable storage medium and a computer program product, and is applied to the technical field of big data, the method comprises the following steps: obtaining historical evaluation text information of a target user, and extracting a plurality of first text keywords from the historical evaluation text information; based on the first text keyword and keyword tags of a plurality of preset evaluation texts, screening an evaluation text set corresponding to each first text keyword from the plurality of preset evaluation texts; for an evaluation text set corresponding to each first text keyword, screening a target evaluation text from the evaluation text set according to an effective evaluation degree label of each evaluation text in the evaluation text set; and generating a service evaluation text according to the target evaluation text. By adopting the method, the obtaining accuracy of the service evaluation text can be improved.

Description

Technical Field

[0001] The present application relates to the field of big data technology, and in particular to a method, apparatus, computer equipment, computer-readable storage medium, and computer program product for generating service evaluation text. Background Art

[0002] With the development of big data, the services provided to users are becoming more and more diversified. In order to make the services provided more in line with user needs, it is usually necessary to collect user feedback on the services after providing services to users, so as to optimize the services based on the user feedback. Therefore, there is an urgent need for a method to obtain user feedback and evaluation.

[0003] At present, it is usually achieved by obtaining evaluation scores from users based on their subjective feelings about the service effects. However, such evaluation scores can reflect little information and cannot provide specific optimization directions for the services. In order to increase the information that can be reflected in the user evaluation, if the specific reasons for the user evaluation scores are further obtained, this will consume the user's patience and may result in less effective information being obtained, resulting in lower accuracy in obtaining the service evaluation text. Summary of the invention

[0004] Based on this, it is necessary to provide a service evaluation text generation method, device, computer equipment, computer-readable storage medium and computer program product that can improve the accuracy of obtaining service evaluation text in response to the above technical problems.

[0005] In a first aspect, the present application provides a method for generating a service evaluation text, comprising:

[0006] Acquire historical evaluation text information of a target user, and extract a plurality of first text keywords from the historical evaluation text information;

[0007] Based on the first text keyword and keyword tags of a plurality of preset evaluation texts, screening an evaluation text set corresponding to each of the first text keywords from the plurality of preset evaluation texts;

[0008] For each evaluation text set corresponding to the first text keyword, selecting a target evaluation text from the evaluation text set according to the effective evaluation degree label of each evaluation text in the evaluation text set;

[0009] A service evaluation text is generated according to the target evaluation text.

[0010] In one of the embodiments, after generating a service evaluation text according to the target evaluation text, the method further includes: obtaining an evaluation content text corresponding to the service evaluation text, and extracting a second text keyword of the target evaluation text from the service evaluation text and the evaluation content text; evaluating the effective evaluation degree of the target evaluation text based on the evaluation content text to obtain an effective evaluation value of the target evaluation text; updating keyword labels of the multiple preset evaluation texts according to the second text keywords of the target evaluation text, and updating the effective evaluation degree labels of the evaluation texts in the evaluation text set according to the effective evaluation value of the target evaluation text.

[0011] In one of the embodiments, the target evaluation text is evaluated for its effective evaluation degree based on the evaluation content text to obtain the effective evaluation value of the target evaluation text, including: obtaining the information complexity of the evaluation content text, and obtaining the association relationship between the evaluation content text and the corresponding preset service; based on the information complexity and the association relationship, the target evaluation text is evaluated for its effective evaluation degree to obtain the effective evaluation value of the target evaluation text.

[0012] In one of the embodiments, the keyword labels of the multiple preset evaluation texts are updated according to the second text keyword of the target evaluation text, and the effective evaluation degree label of each evaluation text in the evaluation text set is updated according to the effective evaluation value of the target evaluation text, including: if the target evaluation text does not meet the update condition, then execute the superposition step: when the updated evaluation content text corresponding to the service evaluation text is detected, the updated evaluation content text is superimposed with the service evaluation text to obtain the superimposed text; if the superimposed text does not meet the update condition, then return to execute the superimposition step; if the superimposed text meets the update condition, then according to the second text keyword of the superimposed text, the keyword labels of the multiple preset evaluation texts are updated, and according to the effective evaluation value of the superimposed text, the effective evaluation degree label of each evaluation text in the evaluation text set is updated.

[0013] In one embodiment, the keyword tags of the multiple preset evaluation texts are updated according to the second text keyword of the superimposed text, including: obtaining a first word frequency of a third text keyword in the second text keyword of the superimposed text, and obtaining a second word frequency of a fourth text keyword in the second text keyword of the superimposed text, wherein the third text keyword is a text keyword in the second text keyword of the superimposed text that corresponds to a keyword tag that does not belong to the multiple preset evaluation texts, and the fourth text keyword is a text keyword in the second text keyword of the superimposed text that corresponds to a keyword tag that belongs to the multiple preset evaluation texts; if the first word frequency corresponding to the third text keyword is greater than a first preset word frequency threshold, the keyword tag corresponding to the third text keyword is added to the keyword tags of the multiple preset evaluation texts; if the second word frequency corresponding to the fourth text keyword is less than the second preset word frequency threshold, the keyword tag corresponding to the fourth text keyword in the keyword tags of the multiple preset evaluation texts is deleted.

[0014] In one of the embodiments, the effective evaluation degree labels of the evaluation texts in the evaluation text set are updated according to the effective evaluation value of the superimposed text, including: obtaining the degree of difference between the effective evaluation value of the superimposed text and the effective evaluation degree labels of each evaluation text in the evaluation text set; if there is a target difference degree that meets a preset difference condition in each of the difference degrees, the effective evaluation degree labels of the evaluation texts in the evaluation text set corresponding to the target difference degree are updated according to the effective evaluation value of the superimposed text.

[0015] In one of the embodiments, the method further includes: obtaining a first text quantity of user portraits corresponding to each text in the superimposed text and belonging to the same user portrait of the target user, and obtaining a second text quantity of abnormal text in the superimposed text, wherein the difference between the text evaluation information corresponding to the abnormal text and the text evaluation information corresponding to the historical evaluation text information is greater than a preset difference threshold; based on the first text quantity and the second text quantity, generating a degree of influence of the superimposed text on the service evaluation text; if the degree of influence is greater than the preset degree threshold, determining that the superimposed text meets the update condition; if the degree of influence is not greater than the preset degree threshold, determining that the superimposed text does not meet the update condition.

[0016] In a second aspect, the present application also provides a service evaluation text generation device, comprising:

[0017] An extraction module, used to obtain historical evaluation text information of a target user, and extract a plurality of first text keywords from the historical evaluation text information;

[0018] A first screening module, configured to screen a set of evaluation texts corresponding to each of the first text keywords from the plurality of preset evaluation texts based on the first text keywords and keyword tags of the plurality of preset evaluation texts;

[0019] A second screening module is used for screening target evaluation texts from the evaluation text set according to the effective evaluation degree label of each evaluation text in the evaluation text set for each evaluation text in the evaluation text set corresponding to the first text keyword;

[0020] A generating module is used to generate a service evaluation text according to the target evaluation text.

[0021] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0022] Acquire historical evaluation text information of a target user, and extract a plurality of first text keywords from the historical evaluation text information;

[0023] Based on the first text keyword and keyword tags of a plurality of preset evaluation texts, screening an evaluation text set corresponding to each of the first text keywords from the plurality of preset evaluation texts;

[0024] For each evaluation text set corresponding to the first text keyword, selecting a target evaluation text from the evaluation text set according to the effective evaluation degree label of each evaluation text in the evaluation text set;

[0025] A service evaluation text is generated according to the target evaluation text.

[0026] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the following steps are implemented:

[0027] Acquire historical evaluation text information of a target user, and extract a plurality of first text keywords from the historical evaluation text information;

[0028] Based on the first text keyword and keyword tags of a plurality of preset evaluation texts, screening an evaluation text set corresponding to each of the first text keywords from the plurality of preset evaluation texts;

[0029] For each evaluation text set corresponding to the first text keyword, selecting a target evaluation text from the evaluation text set according to the effective evaluation degree label of each evaluation text in the evaluation text set;

[0030] A service evaluation text is generated according to the target evaluation text.

[0031] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

[0032] Acquire historical evaluation text information of a target user, and extract a plurality of first text keywords from the historical evaluation text information;

[0033] Based on the first text keyword and keyword tags of a plurality of preset evaluation texts, screening an evaluation text set corresponding to each of the first text keywords from the plurality of preset evaluation texts;

[0034] For each evaluation text set corresponding to the first text keyword, selecting a target evaluation text from the evaluation text set according to the effective evaluation degree label of each evaluation text in the evaluation text set;

[0035] A service evaluation text is generated according to the target evaluation text.

[0036] The above-mentioned service evaluation text generation method, device, computer equipment, computer-readable storage medium and computer program product obtain the historical evaluation text information of the target user and extract multiple first text keywords from the historical evaluation text information; based on the first text keyword and the keyword tags of multiple preset evaluation texts, select the evaluation text set corresponding to each first text keyword from the multiple preset evaluation texts; for the evaluation text set corresponding to each first text keyword, select the target evaluation text from the evaluation text set according to the effective evaluation degree label of each evaluation text in the evaluation text set; generate the service evaluation text according to the target evaluation text, and through two screening processes of the historical evaluation text information, ensure that the selected target evaluation text corresponds to the historical evaluation text information, which can reflect more effective evaluation information, and thus also ensure that the service evaluation text corresponding to the target evaluation text can obtain more effective information, that is, improve the accuracy of obtaining the service evaluation text. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0038] Figure 1 An application environment diagram of a method for generating service evaluation text in one embodiment;

[0039] Figure 2A schematic diagram of a flow chart of a method for generating a service evaluation text in one embodiment;

[0040] Figure 3 A flowchart of the steps of generating a delay time period corresponding to a smart contract in one embodiment;

[0041] Figure 4 A schematic diagram of a flow chart of a step of generating a delay time period corresponding to a smart contract according to a first risk assessment result in an embodiment;

[0042] Figure 5 A structural block diagram of a service evaluation text generating device in one embodiment;

[0043] Figure 6 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0044] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail 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 used to limit the present application.

[0045] It should be noted that the user information (including but not limited to historical evaluation text information, preset evaluation text, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the acquisition, transmission, storage, use and processing of relevant data are in compliance with the relevant provisions of national laws and regulations. For content pushed to users (for example, service evaluation texts, etc.), users can refuse or conveniently refuse content push, etc. In the embodiments of this application, certain software, components, models and other existing solutions in the industry may be mentioned, and they should be regarded as exemplary. Their purpose is only to illustrate the feasibility of the implementation of the technical solution of this application, but it does not mean that the applicant has or will necessarily use the solution.

[0046] The service evaluation text generation method provided in the embodiment of the present application can be applied to Figure 1In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or it can be placed on the cloud or other network servers. The server 104 obtains the historical evaluation text information of the target user, and extracts multiple first text keywords from the historical evaluation text information; based on the first text keyword and the keyword tags of multiple preset evaluation texts, the evaluation text set corresponding to each first text keyword is screened from multiple preset evaluation texts; for the evaluation text set corresponding to each first text keyword, according to the effective evaluation degree label of each evaluation text in the evaluation text set, the target evaluation text is screened from the evaluation text set; the server 104 generates a service evaluation text according to the target evaluation text, and pushes the service evaluation text to the terminal 102. Among them, the terminal 102 can be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, Internet of Things devices and portable wearable devices, and the Internet of Things devices can be smart speakers, smart TVs, smart air conditioners, smart car-mounted devices, projection devices, etc. Portable wearable devices can be smart watches, smart bracelets, head-mounted devices, etc. The head mounted device may be a virtual reality (VR) device, an augmented reality (AR) device, smart glasses, etc. The server 104 may be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0047] In an exemplary embodiment, Figure 2 As shown, a method for generating service evaluation text is provided, and the method is applied to Figure 1 The server 104 in the example is used as an example to illustrate, including the following steps 202 to 208. Among them:

[0048] Step 202: Obtain historical evaluation text information of the target user, and extract a plurality of first text keywords from the historical evaluation text information.

[0049] Among them, the historical evaluation text information in step 202 is the evaluation text information of the target user for the preset service. The preset service is a pre-set service that requires the collection of service evaluation text. The preset service can be a service provided by a certain service item or a service provided by a certain product. The historical evaluation text information can be evaluation content text information, for example, content text similar to a film review, or it can be evaluation question text information and rating information and / or option information corresponding to the evaluation question text information. For example, similar to a questionnaire, the questions in the questionnaire are the evaluation question text information, the ratings corresponding to the questions in the questionnaire are the rating information corresponding to the evaluation question text information, and the options corresponding to the questions in the questionnaire are the option information corresponding to the evaluation question text information. It can also be a combination of evaluation content text information, evaluation question text information and rating information corresponding to the evaluation question text information, which is not limited here.

[0050] Among them, the target user in step 202 is a user who has used the preset service, and the number of target users can be single or multiple. In the case of multiple target users, the target users can be multiple users belonging to the same user profile, or multiple users belonging to different user profiles, without limitation here. The historical evaluation text information of the target user can be the historical evaluation text information under single or multiple evaluations of multiple target users, or the historical evaluation text information under multiple or single evaluations of a single target user, without limitation here.

[0051] Exemplarily, obtaining the historical evaluation text information of the target user includes: querying the historical evaluation text information of the target user from a preset evaluation library.

[0052] As one embodiment, in response to a service evaluation instruction of a preset service, a target user is selected from users who have used the preset service, and historical evaluation text information of the target user is queried. Specifically, the target user is selected from users who have used the preset service, including: selecting all users who have used the preset service as target users, or selecting all users who have used the preset service and are continuous users of the preset service as target users.

[0053] Exemplarily, extracting multiple first text keywords from the historical evaluation text information includes: obtaining word frequency information of each word in the historical evaluation text information, and selecting multiple first text keywords from each word in the historical evaluation text information according to the word frequency information of each word in the historical evaluation text information.

[0054] As an embodiment, word frequency information includes word frequency; obtaining word frequency information of each word in the historical evaluation text information, including: for each word in the historical evaluation text information, determining the word frequency by the ratio of the number of times the word appears in the historical evaluation text information to the number of all words in the historical evaluation text information; selecting multiple first text keywords from each word in the historical evaluation text information based on the word frequency information of each word in the historical evaluation text information, including: selecting the words whose word frequencies corresponding to each word in the historical evaluation text information meet the preset word frequency conditions as the first text keywords, wherein the preset word frequency condition can be greater than a preset word frequency threshold, and the preset word frequency condition can also be a preset number belonging to the top column under the word frequency sorting, for example, selecting the word frequencies top10, top3, etc.

[0055] As another embodiment, the word frequency information includes inverse document frequency and word frequency, and the word frequency information of each word in the historical evaluation text information is obtained, including: for each word in the historical evaluation text information, the logarithm of the ratio between the number of all texts and the number of texts containing the word plus 1 is determined as the inverse document frequency of the word; according to the word frequency information of each word in the historical evaluation text information, multiple first text keywords are selected from each word in the historical evaluation text information, including: for each word in the historical evaluation text information, the product between the word frequency and the inverse document frequency of the word is obtained, and the product is determined as the frequency merging value of the word, and the word whose frequency merging value corresponding to each word in the historical evaluation text information meets the preset frequency merging condition is selected as the first text keyword, wherein the preset frequency merging condition can be greater than a preset frequency merging threshold, and the preset frequency merging condition can also be a preset number belonging to the top column when sorted by the frequency merging value, for example, selecting the frequency merging values ​​top10, top3, etc.

[0056] In this way, the word frequency and the inverse document frequency are used as the basis for selecting the first text keyword, so that the first text keyword can reflect most of the evaluation tendency words of the target user.

[0057] As another embodiment, the word frequency information includes inverse document frequency and word frequency. The word frequency information includes inverse document frequency and word frequency. For each word in the historical evaluation text information, the inverse document frequency of the word is weighted by a first preset weight to obtain an inverse document weighted frequency. The word frequency of the word is weighted by a second preset weight to obtain a weighted word frequency. The inverse document weighted frequency and the weighted word frequency of the word are aggregated to obtain a combined frequency value of the word, wherein the aggregation method includes summation or product.

[0058] Step 204 , based on the first text keyword and the keyword tags of the plurality of preset evaluation texts, a set of evaluation texts corresponding to each first text keyword is screened from the plurality of preset evaluation texts.

[0059] The preset evaluation text in step 204 is a preset evaluation question text and / or evaluation content text of a user who has used the preset service. The evaluation text set is the evaluation content of the preset service by the user who has used the preset service.

[0060] Exemplarily, step 204 includes: obtaining keyword tags of multiple preset evaluation texts, and for each first text keyword, based on the matching degree between the first text keyword and the keyword tags of the multiple preset evaluation texts, filtering the evaluation text set corresponding to the first text keyword from the multiple preset evaluation texts.

[0061] As an embodiment, obtaining keyword tags of multiple preset evaluation texts includes: for each preset evaluation text, obtaining question information corresponding to the preset evaluation text, extracting keywords from the question information corresponding to the preset evaluation text, and obtaining keyword tags of the preset evaluation text.

[0062] It can be understood that when the preset evaluation text corresponds to the question information, the preset evaluation text is usually the scoring content or option selection content for the question information. Therefore, in this case, the question information corresponding to the preset evaluation text is used as the basis for extracting keywords, thereby ensuring the accuracy of the keyword labels extracted from the preset evaluation text.

[0063] As another embodiment, obtaining keyword tags of a plurality of preset evaluation texts includes: for each preset evaluation text, extracting keywords from the preset evaluation text to obtain the keyword tag of the preset evaluation text.

[0064] It is understandable that when there is no corresponding question information in the preset evaluation text, the preset evaluation text is usually an evaluation feedback text for the preset service. Therefore, in this case, the preset evaluation text is used as the basis for extracting keywords, ensuring the accuracy of the keyword tags extracted from the preset evaluation text.

[0065] As one embodiment, based on the matching degrees between the first text keyword and the keyword tags of multiple preset evaluation texts, a set of evaluation texts corresponding to the first text keyword is screened from multiple preset evaluation texts, including: obtaining the matching degrees between the first text keyword and the keyword tags of multiple preset evaluation texts, and screening the evaluation text set corresponding to the first text keyword whose matching degrees are greater than a preset matching degree threshold from the multiple preset evaluation texts.

[0066] Furthermore, obtaining the degree of match between the first text keyword and the keyword tags of multiple preset evaluation texts includes: determining the similarity between the first text keyword and the keyword tags of multiple preset evaluation texts, and determining the similarity between the first text keyword and the keyword tags of multiple preset evaluation texts as the degree of match between the first text keyword and the keyword tags of multiple preset evaluation texts.

[0067] Step 206 , for the evaluation text set corresponding to each first text keyword, select a target evaluation text from the evaluation text set according to the effective evaluation degree label of each evaluation text in the evaluation text set.

[0068] The effective evaluation degree label in step 206 is used to characterize the effectiveness of the information in the evaluation text for the preset service.

[0069] Exemplarily, step 206 includes: for the evaluation text set corresponding to each first text keyword, obtaining the effective evaluation degree label of each evaluation text in the evaluation text set, and screening the target evaluation text whose effective evaluation value represented by the corresponding effective evaluation degree label meets the preset evaluation value condition from the evaluation text set.

[0070] Optionally, the specific implementation method of obtaining the effective evaluation level label of each evaluation text in the evaluation text set can refer to the specific implementation steps of the following step 304, which will not be described in detail here.

[0071] Among them, the preset evaluation value condition can be greater than a preset evaluation value threshold, or it can be sorted from large to small among the effective evaluation values ​​represented by the effective evaluation degree labels, and belong to the top preset number, for example, selecting the effective evaluation values ​​top10, top3, etc.

[0072] In this way, it is ensured that the effective evaluation value represented by the effective evaluation degree label of the evaluation text corresponding to the selected target evaluation text is high.

[0073] Step 208, generating a service evaluation text according to the target evaluation text.

[0074] Exemplarily, step 208 includes: aggregating the target evaluation text to obtain the service evaluation text.

[0075] As an embodiment, step 208 includes: when there are multiple target evaluation texts, each target evaluation text is arranged and combined according to the text question type of each target evaluation text to obtain a service evaluation text, wherein specifically, the target evaluation text whose text question type is the ending feedback type is arranged in the latter part of the service evaluation text, for example, questionnaire end feedback, summary feedback, etc. For the target evaluation text whose text question type does not belong to the ending feedback type, the association relationship between the target evaluation texts whose text question type does not belong to the ending feedback type is obtained, and according to the association relationship between the target evaluation texts whose text question type does not belong to the ending feedback type, the target evaluation texts whose text question type does not belong to the ending feedback type are sorted to obtain the service evaluation text.

[0076] In this way, the target evaluation text is reasonably sorted according to the correlation between the text question type and the target evaluation text, so that the generated service evaluation text is logically reasonable in sequence, thereby avoiding the situation where the user's evaluation willingness is reduced when the generated service evaluation text is logically chaotic in sequence, thereby improving the accuracy of obtaining the service evaluation text.

[0077] In the above-mentioned service evaluation text generation method, historical evaluation text information of the target user is obtained, and multiple first text keywords are extracted from the historical evaluation text information; based on the first text keywords and the keyword tags of multiple preset evaluation texts, an evaluation text set corresponding to each first text keyword is screened from multiple preset evaluation texts; for the evaluation text set corresponding to each first text keyword, a target evaluation text is screened from the evaluation text set according to the effective evaluation degree label of each evaluation text in the evaluation text set; based on the target evaluation text, a service evaluation text is generated, and by performing two screening processes on the historical evaluation text information, it is ensured that the selected target evaluation text corresponds to the historical evaluation text information and can reflect more effective evaluation information, thereby also ensuring that the service evaluation text corresponding to the target evaluation text can obtain more effective information, that is, the accuracy of obtaining the service evaluation text is improved.

[0078] It is understandable that there may be a situation where the generated service evaluation text still cannot obtain effective information. After the service evaluation text is generated, the target user and / or other users provide feedback on the generated service evaluation text, and this feedback can be used to optimize the service evaluation text, thereby ensuring that the optimized service evaluation text can obtain effective information. Therefore, a method for optimizing the service evaluation text is urgently needed.

[0079] In an exemplary embodiment, Figure 3As shown, after generating the service evaluation text according to the target evaluation text, the method of optimizing the service evaluation text includes steps 302 to 306. Among them:

[0080] Step 302: Obtain the evaluation content text corresponding to the service evaluation text, and extract the second text keywords of the target evaluation text from the service evaluation text and the evaluation content text.

[0081] The evaluation content text is the feedback content of the service evaluation text on the preset service by the target user or other users who have used the preset service.

[0082] Exemplarily, obtaining the evaluation content text corresponding to the service evaluation text includes: obtaining the evaluation content text corresponding to the service evaluation text sent by a user terminal used by a target user or other users who have used the preset service, or scanning the content filled in by the target user or other users who have used the preset service for the service evaluation text to obtain the evaluation content text corresponding to the service evaluation text.

[0083] Optionally, the specific implementation method of extracting the second text keywords of the target evaluation text from the service evaluation text and the evaluation content text can refer to the specific implementation steps of extracting multiple first text keyword features from the historical evaluation text information in the above step 202, which will not be repeated here.

[0084] Step 304, based on the evaluation content text, the effective evaluation degree of the target evaluation text is evaluated to obtain the effective evaluation value of the target evaluation text.

[0085] As an embodiment, step 304 includes: obtaining the information complexity of the evaluation content text, and obtaining the association relationship between the evaluation content text and the corresponding preset service; based on the information complexity and the association relationship, evaluating the effective evaluation degree of the target evaluation text to obtain the effective evaluation value of the target evaluation text.

[0086] Further, as an embodiment, obtaining the information complexity of the evaluation content text includes: obtaining the number of fields of the evaluation content text and the field types of the evaluation content text, and generating the information complexity of the evaluation content text according to the number of fields and the field types of the evaluation content text, wherein the larger the number of fields of the evaluation content text, the higher the information complexity of the evaluation content text, and the more field types of the evaluation content text, the higher the information complexity of the evaluation content text.

[0087] In this way, the number and types of fields are used as the basis for evaluating the complexity of information. The complexity of information is evaluated from the dimensions of the number and types of fields, which improves the accuracy of evaluating the complexity of information in the evaluation content text.

[0088] It is understandable that there may be a situation where a large part of the evaluation content text is composed of linking words. In this case, the method of using the above-mentioned field quantity and type dimensions to evaluate the information complexity may result in inaccurate evaluation of the information complexity of the evaluation content text.

[0089] In order to overcome the above technical defect that the evaluation content text contains a large amount of linking words, which leads to inaccurate evaluation of the information complexity of the evaluation content text, as another embodiment, obtaining the information complexity of the evaluation content text includes: obtaining the number of each word and the part of speech of each word in the evaluation content text, and generating the information complexity of the evaluation content text according to the number of each word and the part of speech of each word in the evaluation content text, wherein the greater the number of non-linking words in the evaluation content text, the higher the information complexity of the evaluation content text.

[0090] In this way, the parts of speech and the number of words are used as the basis for evaluating the information complexity. The influence of the number of effective words (non-cohesive words) in the evaluation content text on the information complexity of the evaluation content text is taken into consideration, thereby improving the evaluation accuracy of the information complexity of the evaluation content text.

[0091] As one embodiment, obtaining the association relationship between the evaluation content text and the corresponding preset service includes: obtaining the correlation between each word in the evaluation content text and the preset service, and generating the association relationship between the evaluation content text and the corresponding preset service according to the correlation between each word in the evaluation content text and the preset service, wherein the higher the proportion of words in the evaluation content text whose correlation with the preset service is greater than a preset correlation threshold, the stronger the correlation represented by the association relationship between the evaluation content text and the corresponding preset service.

[0092] As an embodiment, the effective evaluation degree of the target evaluation text is evaluated according to the information complexity and the association relationship to obtain the effective evaluation value of the target evaluation text, including: weighting the information complexity of the evaluation content text by a third preset weight to obtain the weighted degree of information complexity; weighting the association relationship of the evaluation content text by a fourth preset weight to obtain the association weighted relationship; aggregating the association weighted relationship and the weighted degree of information complexity to obtain the aggregated information of the evaluation content text, and evaluating the effective evaluation degree of the target evaluation text according to the aggregated information of the evaluation content text to obtain the effective evaluation value of the target evaluation text, wherein the larger the aggregation value corresponding to the aggregated information of the evaluation content text, the larger the effective evaluation value of the target evaluation text, wherein the third preset weight and the fourth preset weight can be set by the user as needed, or can be an empirical value.

[0093] Step 306, updating the keyword labels of multiple preset evaluation texts according to the second text keyword of the target evaluation text, and updating the effective evaluation degree labels of the evaluation texts in the evaluation text set according to the effective evaluation value of the target evaluation text.

[0094] As an embodiment, keyword tags of multiple preset evaluation texts are updated according to the second text keywords of the target evaluation text, including: updating the keyword tags of multiple preset evaluation texts according to the difference between the second text keywords of the target evaluation text and the keyword tags of the multiple preset evaluation texts.

[0095] As an embodiment, the effective evaluation degree labels of the evaluation texts in the evaluation text set are updated according to the effective evaluation value of the target evaluation text, including: updating the effective evaluation degree labels of the evaluation texts in the evaluation text set to labels whose corresponding effective evaluation values ​​are consistent with the effective evaluation value of the target evaluation text.

[0096] In this embodiment, by obtaining the evaluation content text corresponding to the service evaluation text, the second text keywords of the target evaluation text are extracted from the service evaluation text and the evaluation content text; based on the evaluation content text, the target evaluation text is evaluated for its effective evaluation degree to obtain an effective evaluation value of the target evaluation text; according to the second text keywords of the target evaluation text, the keyword tags of multiple preset evaluation texts are updated, and, according to the effective evaluation value of the target evaluation text, the effective evaluation degree tags of the evaluation texts in the evaluation text set are updated; based on the feedback (evaluation content text) on the service evaluation text from the target user and / or other users who have used the preset service, the keyword tags of multiple preset evaluation texts are updated, and, the effective evaluation degree tags of the evaluation texts in the evaluation text set are updated, thereby achieving optimization of the service evaluation text, forming a closed-loop logic of evaluation-feedback for the preset service, and improving the optimization accuracy of the service evaluation text.

[0097] It is understandable that the feedback from target users and / or other users on the generated service evaluation text may not actually represent the entire evaluation group. In this case, if the service evaluation text is optimized as long as there is feedback from target users and / or other users on the generated service evaluation text, the feedback from target users and / or other users on the generated service evaluation text may be too extreme, resulting in the adjusted service evaluation text still failing to obtain valid information. Therefore, there is an urgent need for a way to accurately optimize the service evaluation text.

[0098] In an exemplary embodiment, Figure 4As shown, according to the second text keyword of the target evaluation text, the keyword labels of multiple preset evaluation texts are updated, and according to the effective evaluation value of the target evaluation text, the effective evaluation degree label of each evaluation text in the evaluation text set is updated, including steps 308 to 310. Among them:

[0099] If the target evaluation text does not meet the update condition, step 308 is executed to perform the superposition step: when the updated evaluation content text corresponding to the service evaluation text is detected, the updated evaluation content text is superimposed on the service evaluation text to obtain the superimposed text.

[0100] As an embodiment, the update condition may be that the number of evaluation users corresponding to the target evaluation text is greater than a first preset number threshold.

[0101] In this way, when the number of users who provide feedback on the service evaluation text is large enough, that is, when the number of feedback samples can affect the service evaluation text generated from the historical evaluation text information of the target user, the service evaluation text can be updated, thereby avoiding the situation where a single sample causes the update of the service evaluation text to actually have a negative effect on the service evaluation text, thereby improving the optimization accuracy of the service evaluation text.

[0102] As another embodiment, the update condition may also be that the number of evaluation users corresponding to the target evaluation text and belonging to the same user profile as the target user is greater than a second preset number threshold.

[0103] In this way, it is further ensured that the optimization basis is the target evaluation text generated by users who belong to the same user profile as the target user, and the updated service evaluation text is further made to fit the user profile of the target user, thereby improving the optimization accuracy of the service evaluation text.

[0104] If the superimposed text does not meet the update condition, the process returns to step 308 .

[0105] If the superimposed text meets the update conditions, step 310 is executed to update the keyword labels of multiple preset evaluation texts according to the second text keyword of the superimposed text, and to update the effective evaluation degree label of each evaluation text in the evaluation text set according to the effective evaluation value of the superimposed text.

[0106] As an embodiment, keyword tags of multiple preset evaluation texts are updated according to the second text keyword of the superimposed text, including: selecting a third text keyword and a fourth text keyword from the second text keyword of the superimposed text, wherein the third text keyword is a text keyword in the second text keyword of the superimposed text that corresponds to a keyword tag that does not belong to multiple preset evaluation texts, and the fourth text keyword is a text keyword in the second text keyword of the superimposed text that corresponds to a keyword tag that belongs to multiple preset evaluation texts; adding the keyword tag corresponding to the third text keyword to the keyword tags of multiple preset evaluation texts, and deleting the keyword tag corresponding to the fourth text keyword in the keyword tags of multiple preset evaluation texts.

[0107] It is understandable that when the number of superimposed texts is large, directly adding tags to the keyword labels of the preset evaluation text according to the third text keywords, and deleting tags to the keyword labels of the preset evaluation text according to the fourth text keywords, may lead to extreme cases where the keyword labels of the preset evaluation text that are finally updated are inaccurate.

[0108] As an embodiment, in order to overcome the above-mentioned technical defect that the keyword labels of the preset evaluation text are added directly according to the third text keyword, and the keyword labels of the preset evaluation text are deleted according to the fourth text keyword, which leads to inaccurate keyword labels of the preset evaluation text in extreme cases, the keyword labels of multiple preset evaluation texts are updated according to the second text keyword of the superimposed text, including: obtaining a first word frequency of the third text keyword in the second text keyword of the superimposed text, and obtaining a second word frequency of the fourth text keyword in the second text keyword of the superimposed text, wherein the third text keyword is a text keyword in the second text keyword of the superimposed text that corresponds to a keyword label that does not belong to multiple preset evaluation texts, and the fourth text keyword is a text keyword in the second text keyword of the superimposed text that corresponds to a keyword label that belongs to multiple preset evaluation texts; if the first word frequency corresponding to the third text keyword is greater than the first preset word frequency threshold, the keyword label corresponding to the third text keyword is added to the keyword labels of multiple preset evaluation texts; if the second word frequency corresponding to the fourth text keyword is less than the second preset word frequency threshold, the keyword label corresponding to the fourth text keyword in the keyword labels of multiple preset evaluation texts is deleted.

[0109] The first preset word frequency threshold and the second preset word frequency threshold can be set by the user as needed, or can be empirical values, which are not limited here.

[0110] In this way, reasonable addition of keyword tags of multiple preset evaluation texts is achieved, so that the keyword tags of the updated preset evaluation texts can accurately represent the content that can enable users to provide effective information, thereby improving the accuracy of the keyword tags of the updated preset evaluation texts.

[0111] As one embodiment, the effective evaluation degree labels of the evaluation texts in the evaluation text set are updated according to the effective evaluation values ​​of the superimposed texts, including: obtaining the degree of difference between the effective evaluation values ​​of the superimposed texts and the effective evaluation degree labels of each evaluation text in the evaluation text set; if there is a target difference degree that meets a preset difference condition among the difference degrees, the effective evaluation degree labels of the evaluation texts in the evaluation text set corresponding to the target difference degree are updated according to the effective evaluation values ​​of the superimposed texts.

[0112] Furthermore, the preset difference condition may be greater than a preset difference degree threshold.

[0113] As one embodiment, the effective evaluation degree labels of the evaluation texts corresponding to the target difference degree in the evaluation text set are updated according to the effective evaluation values ​​of the superimposed texts, including: updating the effective evaluation degree labels of the evaluation texts corresponding to the target difference degree in the evaluation text set to labels whose corresponding effective evaluation values ​​are consistent with the effective evaluation values ​​of the superimposed texts, or updating the effective evaluation degree labels of the evaluation texts corresponding to the target difference degree in the evaluation text set to labels whose corresponding effective evaluation values ​​are added to the effective evaluation values ​​of the superimposed texts to the average value of the effective evaluation values ​​corresponding to the effective evaluation degree labels of the evaluation texts corresponding to the target difference degree.

[0114] In this way, when there is a large difference between the superimposed text actually fed back by the target user or the user who has used the preset service and the effective evaluation degree label of the evaluation text, the effective evaluation degree label of the evaluation text is updated, so that the target evaluation text obtained by filtering the effective evaluation degree label of the updated evaluation text can accurately represent the evaluation text of the actual feedback that can obtain effective information, thereby improving the updating accuracy of the service evaluation text.

[0115] Optionally, the method also includes: obtaining a first text quantity of user portraits corresponding to each text in the superimposed text and belonging to the same user portrait of the target user, and obtaining a second text quantity of abnormal text in the superimposed text, wherein the difference between the text evaluation information corresponding to the abnormal text and the text evaluation information corresponding to the historical evaluation text information is greater than a preset difference threshold; if the degree of influence is greater than the preset degree threshold, determining that the superimposed text meets the update condition; if the degree of influence is not greater than the preset degree threshold, determining that the superimposed text does not meet the update condition, specifically, the larger the number of first texts, the greater the degree of influence; the larger the number of second texts, the greater the degree of influence.

[0116] The text evaluation information corresponding to the abnormal text includes the second text keyword and the effective evaluation value corresponding to the abnormal text; the text evaluation information corresponding to the historical evaluation text information includes the first text keyword and the effective evaluation degree label corresponding to the historical evaluation text information. The preset difference threshold is a difference critical value set in advance as needed. The preset degree threshold is an impact degree critical value set in advance as needed.

[0117] Furthermore, the second text quantity of abnormal text in the superimposed text is obtained, including: for each text in the superimposed text, obtaining a first difference between the second text keyword of the text and the first text keyword of the historical evaluation text information, and obtaining a second difference between the effective evaluation value of the text and the evaluation value represented by the effective evaluation degree label of the historical evaluation text information; aggregating the first difference and the second difference to obtain the difference between the text and the historical evaluation text information.

[0118] In this way, the number of first texts corresponding to the texts belonging to the same user portrait in the superimposed text is used as the basis for generating the update conditions, ensuring that the basis for optimizing the service evaluation text is the texts belonging to the same user portrait as the target user, thereby improving the optimization accuracy of the service evaluation text; the number of second texts of abnormal texts in the superimposed text is used as the basis for generating the update conditions, ensuring that large differences in evaluations over time, such as changes in aesthetics, can be captured, thereby improving the optimization accuracy of the service evaluation text.

[0119] In this embodiment, if the target evaluation text does not meet the update condition, the superposition step is executed: when the updated evaluation content text corresponding to the service evaluation text is detected, the updated evaluation content text is superimposed on the service evaluation text to obtain the superimposed text; if the superimposed text does not meet the update condition, the superimposition step is returned to be executed; if the superimposed text meets the update condition, the keyword tags of multiple preset evaluation texts are updated according to the second text keyword of the superimposed text, and the effective evaluation degree label of each evaluation text in the evaluation text set is updated according to the effective evaluation value of the superimposed text, and the conditions for updating the keyword tags of multiple preset evaluation texts and the effective evaluation degree label of each evaluation text in the evaluation text set are set, thereby avoiding the technical defect that the service evaluation text is updated every time there is a change, resulting in a calculation situation in which the finally updated service evaluation text cannot obtain more effective information. Therefore, the reasonable update of the service evaluation text is guaranteed and the update accuracy of the service evaluation text is improved.

[0120] As a detailed embodiment, historical evaluation text information of a target user is obtained, and a plurality of first text keywords are extracted from the historical evaluation text information; based on the first text keywords and keyword tags of a plurality of preset evaluation texts, an evaluation text set corresponding to each first text keyword is screened from a plurality of preset evaluation texts; for the evaluation text set corresponding to each first text keyword, a target evaluation text is screened from the evaluation text set according to the effective evaluation degree tag of each evaluation text in the evaluation text set; a service evaluation text is generated according to the target evaluation text; an evaluation content text corresponding to the service evaluation text is obtained, and a second text keyword of the target evaluation text is extracted from the service evaluation text and the evaluation content text; the information complexity of the evaluation content text is obtained, and the association relationship between the evaluation content text and the corresponding preset service is obtained; based on the information complexity and the association relationship, the effective evaluation degree of the target evaluation text is evaluated to obtain the effective evaluation value of the target evaluation text.

[0121] Furthermore, if the target evaluation text does not meet the update condition, the superposition step is performed: when an updated evaluation content text corresponding to the service evaluation text is detected, the updated evaluation content text is superimposed on the service evaluation text to obtain the superimposed text; the first text quantity of the user portrait corresponding to each text in the superimposed text and the target user belonging to the same user portrait, and the second text quantity of abnormal text in the superimposed text, wherein the difference between the text evaluation information corresponding to the abnormal text and the text evaluation information corresponding to the historical evaluation text information is greater than a preset difference threshold; based on the first text quantity and the second text quantity, the influence degree of the superimposed text on the service evaluation text is generated; if the influence degree is greater than the preset degree threshold, it is determined that the superimposed text meets the update condition; if the influence degree is not greater than the preset degree threshold, it is determined that the superimposed text does not meet the update condition; if the superimposed text does not meet the update condition, return to execute the superimposition step; if the superimposed text meets the update condition, the first word frequency of the third text keyword in the second text keyword of the superimposed text is obtained, and the fourth text keyword is obtained. The second word frequency of this keyword in the second text keyword of the superimposed text, wherein the third text keyword is a text keyword in the second text keyword of the superimposed text that corresponds to a keyword label that does not belong to multiple preset evaluation texts, and the fourth text keyword is a text keyword in the second text keyword of the superimposed text that corresponds to a keyword label that belongs to multiple preset evaluation texts; if the first word frequency corresponding to the third text keyword is greater than the first preset word frequency threshold, the keyword label corresponding to the third text keyword is added to the keyword labels of multiple preset evaluation texts; if the second word frequency corresponding to the fourth text keyword is less than the second preset word frequency threshold, the keyword label corresponding to the fourth text keyword in the keyword labels of multiple preset evaluation texts is deleted; the difference degree between the effective evaluation value of the superimposed text and the effective evaluation degree label of each evaluation text in the evaluation text set is obtained; if there is a target difference degree that meets the preset difference condition in each difference degree, the effective evaluation degree label of the evaluation text in the evaluation text set corresponding to the target difference degree is updated according to the effective evaluation value of the superimposed text.

[0122] In this way, the historical evaluation text information of the target user is obtained, and multiple first text keywords are extracted from the historical evaluation text information; based on the first text keywords and the keyword tags of multiple preset evaluation texts, the evaluation text set corresponding to each first text keyword is screened from the multiple preset evaluation texts; for the evaluation text set corresponding to each first text keyword, the target evaluation text is screened from the evaluation text set according to the effective evaluation degree label of each evaluation text in the evaluation text set; based on the target evaluation text, a service evaluation text is generated, and through two screening processes of the historical evaluation text information, it is ensured that the selected target evaluation text corresponds to the historical evaluation text information, and can reflect more effective evaluation information, thereby also ensuring that the service evaluation text corresponding to the target evaluation text can obtain more effective information, that is, the accuracy of obtaining the service evaluation text is improved.

[0123] Furthermore, based on the feedback (evaluation content text) of the target users and / or other users who have used the preset service on the service evaluation text, the keyword tags of multiple preset evaluation texts are updated, and the effective evaluation degree tags of the evaluation texts in the evaluation text set are updated, so as to optimize the service evaluation text and form a closed-loop logic of evaluation-feedback for the preset service, thereby ensuring that the optimized service evaluation text can obtain more effective information; and setting conditions for updating the keyword tags of multiple preset evaluation texts and the effective evaluation degree tags of each evaluation text in the evaluation text set, thereby avoiding the technical defect that the service evaluation text is updated every time there is a change, resulting in a calculation situation in which the finally updated service evaluation text cannot obtain more effective information. Therefore, the reasonable update of the service evaluation text is guaranteed and the update accuracy of the service evaluation text is improved.

[0124] It should be understood that, although the steps in the flowcharts involved in the above embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above embodiments may include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0125] Based on the same inventive concept, the embodiment of the present application also provides a service evaluation text generation device for implementing the service evaluation text generation method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more service evaluation text generation device embodiments provided below can refer to the limitations of the service evaluation text generation method above, and will not be repeated here.

[0126] In an exemplary embodiment, Figure 5 As shown, a service evaluation text generation device 500 is provided, comprising: an extraction module 502, a first screening module 504, a second screening module 506 and a generation module 508, wherein:

[0127] Extraction module 502, used to obtain historical evaluation text information of a target user, and extract a plurality of first text keywords from the historical evaluation text information;

[0128] A first screening module 504 is used to screen the evaluation text set corresponding to each first text keyword from the plurality of preset evaluation texts based on the first text keyword and the keyword tags of the plurality of preset evaluation texts;

[0129] The second screening module 506 is used to screen target evaluation texts from the evaluation text set according to the effective evaluation degree label of each evaluation text in the evaluation text set for each evaluation text set corresponding to the first text keyword;

[0130] The generating module 508 is used to generate a service evaluation text according to the target evaluation text.

[0131] In one of the embodiments, after generating a service evaluation text based on a target evaluation text, the device also includes an updating module for obtaining an evaluation content text corresponding to the service evaluation text, extracting a second text keyword of the target evaluation text from the service evaluation text and the evaluation content text; evaluating the effective evaluation degree of the target evaluation text based on the evaluation content text to obtain an effective evaluation value of the target evaluation text; updating keyword labels of multiple preset evaluation texts based on the second text keywords of the target evaluation text, and updating the effective evaluation degree labels of the evaluation texts in the evaluation text set based on the effective evaluation value of the target evaluation text.

[0132] In one of the embodiments, the update module is also used to obtain the information complexity of the evaluation content text, and to obtain the association relationship between the evaluation content text and the corresponding preset service; based on the information complexity and the association relationship, the effective evaluation degree of the target evaluation text is evaluated to obtain the effective evaluation value of the target evaluation text.

[0133] In one of the embodiments, the update module is also used to execute the superimposition step if the target evaluation text does not meet the update conditions: when the updated evaluation content text corresponding to the service evaluation text is detected, the updated evaluation content text is superimposed on the service evaluation text to obtain the superimposed text; if the superimposed text does not meet the update conditions, return to execute the superimposition step; if the superimposed text meets the update conditions, the keyword labels of multiple preset evaluation texts are updated according to the second text keyword of the superimposed text, and the effective evaluation degree label of each evaluation text in the evaluation text set is updated according to the effective evaluation value of the superimposed text.

[0134] In one embodiment, the update module is also used to obtain a first word frequency of a third text keyword in a second text keyword of the superimposed text, and to obtain a second word frequency of a fourth text keyword in the second text keyword of the superimposed text, wherein the third text keyword is a text keyword in the second text keyword of the superimposed text that corresponds to a keyword label that does not belong to multiple preset evaluation texts, and the fourth text keyword is a text keyword in the second text keyword of the superimposed text that corresponds to a keyword label that belongs to multiple preset evaluation texts; if the first word frequency corresponding to the third text keyword is greater than a first preset word frequency threshold, the keyword label corresponding to the third text keyword is added to the keyword labels of multiple preset evaluation texts; if the second word frequency corresponding to the fourth text keyword is less than the second preset word frequency threshold, the keyword label corresponding to the fourth text keyword in the keyword labels of multiple preset evaluation texts is deleted.

[0135] In one of the embodiments, the update module is also used to obtain the degree of difference between the effective evaluation value of the superimposed text and the effective evaluation degree label of each evaluation text in the evaluation text set; if there is a target difference degree that meets the preset difference condition among the difference degrees, the effective evaluation degree label of the evaluation text corresponding to the target difference degree in the evaluation text set is updated according to the effective evaluation value of the superimposed text.

[0136] In one of the embodiments, the update module is also used to obtain the first text quantity of the user portrait corresponding to each text in the superimposed text and the target user belonging to the same user portrait, and to obtain the second text quantity of abnormal text in the superimposed text, wherein the difference between the text evaluation information corresponding to the abnormal text and the text evaluation information corresponding to the historical evaluation text information is greater than a preset difference threshold; based on the first text quantity and the second text quantity, generate the degree of influence of the superimposed text on the service evaluation text; if the degree of influence is greater than the preset degree threshold, determine that the superimposed text meets the update condition; if the degree of influence is not greater than the preset degree threshold, determine that the superimposed text does not meet the update condition.

[0137] Each module in the above service evaluation text generation device can be implemented in whole or in part by software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each module.

[0138] In an exemplary embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. Figure 6 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store object information and service information. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for generating a service evaluation text is implemented.

[0139] Those skilled in the art will understand that Figure 6 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0140] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0141] Acquire historical evaluation text information of a target user, and extract a plurality of first text keywords from the historical evaluation text information;

[0142] Based on the first text keyword and the keyword tags of the plurality of preset evaluation texts, screening the evaluation text set corresponding to each first text keyword from the plurality of preset evaluation texts;

[0143] For each evaluation text set corresponding to the first text keyword, according to the effective evaluation degree label of each evaluation text in the evaluation text set, a target evaluation text is selected from the evaluation text set;

[0144] Generate service evaluation text based on the target evaluation text.

[0145] In one embodiment, the processor also implements the following steps when executing the computer program: obtaining the evaluation content text corresponding to the service evaluation text, extracting the second text keywords of the target evaluation text from the service evaluation text and the evaluation content text; based on the evaluation content text, evaluating the effective evaluation degree of the target evaluation text to obtain the effective evaluation value of the target evaluation text; updating the keyword labels of multiple preset evaluation texts according to the second text keywords of the target evaluation text, and updating the effective evaluation degree labels of the evaluation texts in the evaluation text set according to the effective evaluation value of the target evaluation text.

[0146] In one embodiment, when the processor executes the computer program, it also implements the following steps: obtaining the information complexity of the evaluation content text, and obtaining the association relationship between the evaluation content text and the corresponding preset service; based on the information complexity and the association relationship, evaluating the effective evaluation degree of the target evaluation text to obtain the effective evaluation value of the target evaluation text.

[0147] In one embodiment, the processor also implements the following steps when executing the computer program: if the target evaluation text does not meet the update condition, the superposition step is executed: when the updated evaluation content text corresponding to the service evaluation text is detected, the updated evaluation content text is superimposed on the service evaluation text to obtain the superimposed text; if the superimposed text does not meet the update condition, the superimposition step is returned to be executed; if the superimposed text meets the update condition, the keyword labels of multiple preset evaluation texts are updated according to the second text keyword of the superimposed text, and the effective evaluation degree label of each evaluation text in the evaluation text set is updated according to the effective evaluation value of the superimposed text.

[0148] In one embodiment, when the processor executes the computer program, the following steps are also implemented: obtaining a first word frequency of a third text keyword in a second text keyword of the superimposed text, and obtaining a second word frequency of a fourth text keyword in the second text keyword of the superimposed text, wherein the third text keyword is a text keyword in the second text keyword of the superimposed text that corresponds to a keyword tag that does not belong to multiple preset evaluation texts, and the fourth text keyword is a text keyword in the second text keyword of the superimposed text that corresponds to a keyword tag that belongs to multiple preset evaluation texts; if the first word frequency corresponding to the third text keyword is greater than a first preset word frequency threshold, the keyword tag corresponding to the third text keyword is added to the keyword tags of multiple preset evaluation texts; if the second word frequency corresponding to the fourth text keyword is less than the second preset word frequency threshold, the keyword tag corresponding to the fourth text keyword in the keyword tags of multiple preset evaluation texts is deleted.

[0149] In one embodiment, when the processor executes the computer program, the following steps are also implemented: obtaining the degree of difference between the effective evaluation value of the superimposed text and the effective evaluation degree label of each evaluation text in the evaluation text set; if there is a target difference degree that meets the preset difference condition among the difference degrees, the effective evaluation degree label of the evaluation text in the evaluation text set corresponding to the target difference degree is updated according to the effective evaluation value of the superimposed text.

[0150] In one embodiment, when the processor executes the computer program, the following steps are also implemented: obtaining the first text quantity of the user portrait corresponding to each text in the superimposed text and the target user belonging to the same user portrait, and obtaining the second text quantity of abnormal text in the superimposed text, wherein the difference between the text evaluation information corresponding to the abnormal text and the text evaluation information corresponding to the historical evaluation text information is greater than a preset difference threshold; based on the first text quantity and the second text quantity, generating the degree of influence of the superimposed text on the service evaluation text; if the degree of influence is greater than the preset degree threshold, determining that the superimposed text meets the update condition; if the degree of influence is not greater than the preset degree threshold, determining that the superimposed text does not meet the update condition.

[0151] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0152] Acquire historical evaluation text information of a target user, and extract a plurality of first text keywords from the historical evaluation text information;

[0153] Based on the first text keyword and the keyword tags of the plurality of preset evaluation texts, screening the evaluation text set corresponding to each first text keyword from the plurality of preset evaluation texts;

[0154] For each evaluation text set corresponding to the first text keyword, according to the effective evaluation degree label of each evaluation text in the evaluation text set, a target evaluation text is selected from the evaluation text set;

[0155] Generate service evaluation text based on the target evaluation text.

[0156] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: obtaining the evaluation content text corresponding to the service evaluation text, extracting the second text keywords of the target evaluation text from the service evaluation text and the evaluation content text; based on the evaluation content text, evaluating the effective evaluation degree of the target evaluation text to obtain the effective evaluation value of the target evaluation text; updating the keyword labels of multiple preset evaluation texts according to the second text keywords of the target evaluation text, and updating the effective evaluation degree labels of the evaluation texts in the evaluation text set according to the effective evaluation value of the target evaluation text.

[0157] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: obtaining the information complexity of the evaluation content text, and obtaining the association relationship between the evaluation content text and the corresponding preset service; based on the information complexity and the association relationship, evaluating the effective evaluation degree of the target evaluation text to obtain the effective evaluation value of the target evaluation text.

[0158] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: if the target evaluation text does not meet the update condition, the superposition step is executed: when the updated evaluation content text corresponding to the service evaluation text is detected, the updated evaluation content text is superimposed on the service evaluation text to obtain the superimposed text; if the superimposed text does not meet the update condition, the superimposition step is returned to be executed; if the superimposed text meets the update condition, the keyword labels of multiple preset evaluation texts are updated according to the second text keyword of the superimposed text, and the effective evaluation degree label of each evaluation text in the evaluation text set is updated according to the effective evaluation value of the superimposed text.

[0159] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: obtaining a first word frequency of a third text keyword in a second text keyword of the superimposed text, and obtaining a second word frequency of a fourth text keyword in the second text keyword of the superimposed text, wherein the third text keyword is a text keyword in the second text keyword of the superimposed text that corresponds to a keyword tag that does not belong to multiple preset evaluation texts, and the fourth text keyword is a text keyword in the second text keyword of the superimposed text that corresponds to a keyword tag that belongs to multiple preset evaluation texts; if the first word frequency corresponding to the third text keyword is greater than a first preset word frequency threshold, the keyword tag corresponding to the third text keyword is added to the keyword tags of multiple preset evaluation texts; if the second word frequency corresponding to the fourth text keyword is less than the second preset word frequency threshold, the keyword tag corresponding to the fourth text keyword in the keyword tags of multiple preset evaluation texts is deleted.

[0160] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: obtaining the degree of difference between the effective evaluation value of the superimposed text and the effective evaluation degree label of each evaluation text in the evaluation text set; if there is a target difference degree that meets the preset difference condition among the difference degrees, the effective evaluation degree label of the evaluation text in the evaluation text set corresponding to the target difference degree is updated according to the effective evaluation value of the superimposed text.

[0161] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: obtaining the first text quantity of the user portrait corresponding to each text in the superimposed text and the target user belonging to the same user portrait, and obtaining the second text quantity of abnormal text in the superimposed text, wherein the difference between the text evaluation information corresponding to the abnormal text and the text evaluation information corresponding to the historical evaluation text information is greater than a preset difference threshold; based on the first text quantity and the second text quantity, generating the degree of influence of the superimposed text on the service evaluation text; if the degree of influence is greater than the preset degree threshold, determining that the superimposed text meets the update condition; if the degree of influence is not greater than the preset degree threshold, determining that the superimposed text does not meet the update condition.

[0162] In one embodiment, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the following steps:

[0163] Acquire historical evaluation text information of a target user, and extract a plurality of first text keywords from the historical evaluation text information;

[0164] Based on the first text keyword and the keyword tags of the plurality of preset evaluation texts, screening the evaluation text set corresponding to each first text keyword from the plurality of preset evaluation texts;

[0165] For each evaluation text set corresponding to the first text keyword, according to the effective evaluation degree label of each evaluation text in the evaluation text set, a target evaluation text is selected from the evaluation text set;

[0166] Generate service evaluation text based on the target evaluation text.

[0167] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: obtaining the evaluation content text corresponding to the service evaluation text, extracting the second text keywords of the target evaluation text from the service evaluation text and the evaluation content text; based on the evaluation content text, evaluating the effective evaluation degree of the target evaluation text to obtain the effective evaluation value of the target evaluation text; updating the keyword labels of multiple preset evaluation texts according to the second text keywords of the target evaluation text, and updating the effective evaluation degree labels of the evaluation texts in the evaluation text set according to the effective evaluation value of the target evaluation text.

[0168] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: obtaining the information complexity of the evaluation content text, and obtaining the association relationship between the evaluation content text and the corresponding preset service; based on the information complexity and the association relationship, evaluating the effective evaluation degree of the target evaluation text to obtain the effective evaluation value of the target evaluation text.

[0169] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: if the target evaluation text does not meet the update condition, the superposition step is executed: when the updated evaluation content text corresponding to the service evaluation text is detected, the updated evaluation content text is superimposed on the service evaluation text to obtain the superimposed text; if the superimposed text does not meet the update condition, the superimposition step is returned to be executed; if the superimposed text meets the update condition, the keyword labels of multiple preset evaluation texts are updated according to the second text keyword of the superimposed text, and the effective evaluation degree label of each evaluation text in the evaluation text set is updated according to the effective evaluation value of the superimposed text.

[0170] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: obtaining a first word frequency of a third text keyword in a second text keyword of the superimposed text, and obtaining a second word frequency of a fourth text keyword in the second text keyword of the superimposed text, wherein the third text keyword is a text keyword in the second text keyword of the superimposed text that corresponds to a keyword tag that does not belong to multiple preset evaluation texts, and the fourth text keyword is a text keyword in the second text keyword of the superimposed text that corresponds to a keyword tag that belongs to multiple preset evaluation texts; if the first word frequency corresponding to the third text keyword is greater than a first preset word frequency threshold, the keyword tag corresponding to the third text keyword is added to the keyword tags of multiple preset evaluation texts; if the second word frequency corresponding to the fourth text keyword is less than the second preset word frequency threshold, the keyword tag corresponding to the fourth text keyword in the keyword tags of multiple preset evaluation texts is deleted.

[0171] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: obtaining the degree of difference between the effective evaluation value of the superimposed text and the effective evaluation degree label of each evaluation text in the evaluation text set; if there is a target difference degree that meets the preset difference condition among the difference degrees, the effective evaluation degree label of the evaluation text in the evaluation text set corresponding to the target difference degree is updated according to the effective evaluation value of the superimposed text.

[0172] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: obtaining the first text quantity of the user portrait corresponding to each text in the superimposed text and the target user belonging to the same user portrait, and obtaining the second text quantity of abnormal text in the superimposed text, wherein the difference between the text evaluation information corresponding to the abnormal text and the text evaluation information corresponding to the historical evaluation text information is greater than a preset difference threshold; based on the first text quantity and the second text quantity, generating the degree of influence of the superimposed text on the service evaluation text; if the degree of influence is greater than the preset degree threshold, determining that the superimposed text meets the update condition; if the degree of influence is not greater than the preset degree threshold, determining that the superimposed text does not meet the update condition.

[0173] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0174] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above 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 application.

[0175] The above 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 present application. 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 present application shall be subject to the attached claims.

Claims

1. A method for generating a service evaluation text, characterized in that: The method comprises: Acquire historical evaluation text information of a target user, and extract a plurality of first text keywords from the historical evaluation text information; Based on the first text keyword and keyword tags of a plurality of preset evaluation texts, screening an evaluation text set corresponding to each of the first text keywords from the plurality of preset evaluation texts; For each evaluation text set corresponding to the first text keyword, selecting a target evaluation text from the evaluation text set according to the effective evaluation degree label of each evaluation text in the evaluation text set; A service evaluation text is generated according to the target evaluation text.

2. The method according to claim 1, characterized in that After generating the service evaluation text according to the target evaluation text, the method further includes: Acquire the evaluation content text corresponding to the service evaluation text, and extract the second text keyword of the target evaluation text from the service evaluation text and the evaluation content text; Based on the evaluation content text, evaluating the effective evaluation degree of the target evaluation text to obtain the effective evaluation value of the target evaluation text; The keyword labels of the plurality of preset evaluation texts are updated according to the second text keyword of the target evaluation text, and the effective evaluation degree labels of the evaluation texts in the evaluation text set are updated according to the effective evaluation value of the target evaluation text.

3. The method according to claim 2, characterized in that The step of evaluating the effective evaluation degree of the target evaluation text based on the evaluation content text to obtain the effective evaluation value of the target evaluation text includes: Obtaining the information complexity of the evaluation content text, and obtaining the association relationship between the evaluation content text and the corresponding preset service; According to the information complexity and the association relationship, the effective evaluation degree of the target evaluation text is evaluated to obtain the effective evaluation value of the target evaluation text.

4. The method according to claim 2, characterized in that: The updating of the keyword labels of the plurality of preset evaluation texts according to the second text keyword of the target evaluation text, and the updating of the effective evaluation degree label of each evaluation text in the evaluation text set according to the effective evaluation value of the target evaluation text, include: If the target evaluation text does not meet the update condition, a superposition step is performed: when an updated evaluation content text corresponding to the service evaluation text is detected, the updated evaluation content text is superimposed on the service evaluation text to obtain a superimposed text; If the superimposed text does not meet the update condition, return to execute the superimposed step; If the superimposed text meets the update condition, the keyword labels of the multiple preset evaluation texts are updated according to the second text keyword of the superimposed text, and the effective evaluation degree label of each evaluation text in the evaluation text set is updated according to the effective evaluation value of the superimposed text.

5. The method according to claim 4, characterized in that The updating of the keyword labels of the plurality of preset evaluation texts according to the second text keyword of the superimposed text includes: Obtaining a first word frequency of a third text keyword in a second text keyword of the superimposed text, and obtaining a second word frequency of a fourth text keyword in the second text keyword of the superimposed text, wherein the third text keyword is a text keyword in the second text keyword of the superimposed text that corresponds to a keyword label that does not belong to the plurality of preset evaluation texts, and the fourth text keyword is a text keyword in the second text keyword of the superimposed text that corresponds to a keyword label that belongs to the plurality of preset evaluation texts; If the first word frequency corresponding to the third text keyword is greater than the first preset word frequency threshold, adding the keyword tag corresponding to the third text keyword to the keyword tags of the plurality of preset evaluation texts; If the second word frequency corresponding to the fourth text keyword is less than a second preset word frequency threshold, the keyword tag corresponding to the fourth text keyword among the keyword tags of the plurality of preset evaluation texts is deleted.

6. The method according to claim 4, characterized in that The updating of the effective evaluation degree labels of the evaluation texts in the evaluation text set according to the effective evaluation values ​​of the superimposed texts comprises: Obtaining the difference between the effective evaluation value of the superimposed text and the effective evaluation degree label of each evaluation text in the evaluation text set; If there is a target difference degree that meets the preset difference condition among the difference degrees, the effective evaluation degree label of the evaluation text corresponding to the target difference degree in the evaluation text set is updated according to the effective evaluation value of the superimposed text.

7. The method according to claim 4, characterized in that The method further comprises: Obtaining the number of first texts of the user portrait corresponding to each text in the superimposed text and the target user belonging to the same user portrait, and obtaining the number of second texts of abnormal texts in the superimposed text, wherein the difference between the text evaluation information corresponding to the abnormal text and the text evaluation information corresponding to the historical evaluation text information is greater than a preset difference threshold; Based on the first text quantity and the second text quantity, generating the influence degree of the superimposed text on the service evaluation text; If the impact degree is greater than a preset degree threshold, determining that the superimposed text meets the update condition; If the impact degree is not greater than the preset degree threshold, it is determined that the superimposed text does not meet the update condition.

8. A service evaluation text generation device, characterized in that: The device comprises: An extraction module, used to obtain historical evaluation text information of a target user, and extract a plurality of first text keywords from the historical evaluation text information; A first screening module, configured to screen a set of evaluation texts corresponding to each of the first text keywords from the plurality of preset evaluation texts based on the first text keywords and keyword tags of the plurality of preset evaluation texts; A second screening module is used for screening target evaluation texts from the evaluation text set according to the effective evaluation degree label of each evaluation text in the evaluation text set for each evaluation text in the evaluation text set corresponding to the first text keyword; A generating module is used to generate a service evaluation text according to the target evaluation text.

9. The device according to claim 8, characterized in that After generating the service evaluation text according to the target evaluation text, the device further comprises an updating module for acquiring the evaluation content text corresponding to the service evaluation text, and extracting the second text keyword of the target evaluation text from the service evaluation text and the evaluation content text; Based on the evaluation content text, evaluating the effective evaluation degree of the target evaluation text to obtain the effective evaluation value of the target evaluation text; The keyword labels of the plurality of preset evaluation texts are updated according to the second text keyword of the target evaluation text, and the effective evaluation degree labels of the evaluation texts in the evaluation text set are updated according to the effective evaluation value of the target evaluation text.

10. The device according to claim 9, characterized in that The updating module is also used to obtain the information complexity of the evaluation content text, and to obtain the association relationship between the evaluation content text and the corresponding preset service; According to the information complexity and the association relationship, the effective evaluation degree of the target evaluation text is evaluated to obtain the effective evaluation value of the target evaluation text.

11. The device according to claim 9, characterized in that The updating module is further configured to perform a superposition step if the target evaluation text does not meet the updating condition: when an updated evaluation content text corresponding to the service evaluation text is detected, superimpose the updated evaluation content text with the service evaluation text to obtain a superimposed text; If the superimposed text does not meet the update conditions, return to execute the superimposition step; if the superimposed text meets the update conditions, the keyword labels of the multiple preset evaluation texts are updated according to the second text keyword of the superimposed text, and the effective evaluation degree label of each evaluation text in the evaluation text set is updated according to the effective evaluation value of the superimposed text.

12. The device according to claim 11, characterized in that The update module is also used to obtain a first word frequency of a third text keyword in the second text keyword of the superimposed text, and to obtain a second word frequency of a fourth text keyword in the second text keyword of the superimposed text, wherein the third text keyword is a text keyword in the second text keyword of the superimposed text that corresponds to a keyword label that does not belong to the multiple preset evaluation texts, and the fourth text keyword is a text keyword in the second text keyword of the superimposed text that corresponds to a keyword label that belongs to the multiple preset evaluation texts; if the first word frequency corresponding to the third text keyword is greater than a first preset word frequency threshold, the keyword label corresponding to the third text keyword is added to the keyword labels of the multiple preset evaluation texts; if the second word frequency corresponding to the fourth text keyword is less than the second preset word frequency threshold, the keyword label corresponding to the fourth text keyword in the keyword labels of the multiple preset evaluation texts is deleted.

13. The device according to claim 11, characterized in that The updating module is also used to obtain the degree of difference between the effective evaluation value of the superimposed text and the effective evaluation degree label of each evaluation text in the evaluation text set; if there is a target difference degree that meets the preset difference condition among the difference degrees, the effective evaluation degree label of the evaluation text corresponding to the target difference degree in the evaluation text set is updated according to the effective evaluation value of the superimposed text.

14. The device according to claim 11, characterized in that The update module is also used to obtain the first text quantity of the user portrait corresponding to each text in the superimposed text and the target user belonging to the same user portrait, and to obtain the second text quantity of abnormal text in the superimposed text, wherein the difference between the text evaluation information corresponding to the abnormal text and the text evaluation information corresponding to the historical evaluation text information is greater than a preset difference threshold; based on the first text quantity and the second text quantity, generate the influence of the superimposed text on the service evaluation text; if the influence is greater than the preset degree threshold, determine that the superimposed text meets the update condition; if the influence is not greater than the preset degree threshold, determine that the superimposed text does not meet the update condition.

15. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

16. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

17. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images