A sentiment analysis method and device based on gas station customer evaluations

Abstract

The present invention relates to the technical field of gas stations, and discloses a sentiment analysis method and device based on customer evaluations of gas stations. The method includes establishing a sentiment dictionary based on at least two evaluation data corresponding to a gas station, and determining a sentiment score corresponding to each evaluation data according to the sentiment dictionary; determining a sentiment tendency corresponding to each evaluation data based on each evaluation data and the corresponding sentiment score; classifying all the evaluation data based on the sentiment tendency to obtain at least three categories of labeled data, and counting the number of occurrences of at least two preset dimensional words in each category of labeled data to obtain statistical data; generating an analysis result corresponding to the gas station based on the sentiment tendency and the statistical data. Through the sentiment dictionary established from the customer evaluations of the gas station, the present application accurately identifies the sentiment tendency in the customer evaluations, improves the accuracy of sentiment analysis of customer evaluations, and is conducive to improving the service quality of gas stations and enhancing customer satisfaction.

Description

Technical Field

[0001] This application relates to the technical field of gas stations, and particularly to a sentiment analysis method and device based on customer evaluations of gas stations. Background Art

[0002] In the current gas station service market, customer feedback is crucial for improving service quality. To facilitate customers and managers in obtaining evaluation information about gas stations more intuitively and quickly, thereby enhancing service levels, competitiveness, and customer experience, the adoption of a gas station evaluation system is particularly important.

[0003] However, existing evaluation systems usually only provide basic text feedback and fail to deeply explore the sentiment tendencies in the evaluations. As a result, information collection and analysis are insufficient, and customer feedback is not effectively utilized, making it impossible to better improve gas station services or more effectively enhance customer satisfaction. Summary of the Invention

[0004] To address the problems of insufficient information collection and analysis, ineffective utilization of customer feedback, and the resulting inability to more effectively enhance customer satisfaction mentioned above, embodiments of this application provide a sentiment analysis method and device based on customer evaluations of gas stations, and the technical solutions are as follows:

[0005] In a first aspect, embodiments of this application provide a sentiment analysis method based on customer evaluations of gas stations, including:

[0006] Establish a sentiment dictionary based on at least two evaluation data corresponding to a gas station, and determine a sentiment score corresponding to each evaluation data according to the sentiment dictionary; wherein, the sentiment dictionary includes at least two sentiment texts generated from the evaluation data and the sentiment scores corresponding to each sentiment text;

[0007] Based on each evaluation data and the corresponding sentiment score, determine the sentiment tendency corresponding to each evaluation data;

[0008] Classify all evaluation data based on the sentiment tendency to obtain at least three categories of labeled data, and count the number of occurrences of at least two preset dimensional words in each category of labeled data to obtain statistical data;

[0009] Generate an analysis result corresponding to the gas station based on the sentiment tendency and the statistical data.

[0010] In an optional solution of the first aspect, establishing a sentiment dictionary based on at least two evaluation data corresponding to a gas station includes:

[0011] Extract at least two sentiment texts from at least two evaluation data corresponding to the gas station;

[0012] Perform annotation processing on all sentiment texts, and calculate the sentiment scores corresponding to each sentiment text based on the results of the annotation processing;

[0013] Based on all sentiment texts and the sentiment scores corresponding to each sentiment text, establish a sentiment dictionary.

[0014] In another alternative solution of the first aspect, performing annotation processing on all sentiment texts, and calculating the sentiment scores corresponding to each sentiment text, includes:

[0015] Perform annotation processing on all sentiment texts, and determine the numerical range of the sentiment scores corresponding to each sentiment text based on the results of the annotation processing;

[0016] Count the number of the same sentiment texts to obtain a first value;

[0017] Count the number of all sentiment texts to obtain a second value;

[0018] Calculate the sentiment scores corresponding to each sentiment text based on the first value, the second value, and the numerical range.

[0019] In another alternative solution of the first aspect, calculating the sentiment scores corresponding to each sentiment text based on the first value, the second value, and the numerical range, includes:

[0020] Determine the intensity coefficient according to the ratio of the first value to the second value;

[0021] Calculate the sentiment scores corresponding to each sentiment text based on the intensity coefficient and the numerical range.

[0022] In another alternative solution of the first aspect, based on each evaluation data and the corresponding sentiment score, determine the sentiment tendency corresponding to each evaluation data, including:

[0023] Preprocess each evaluation data to obtain target data, and generate an initial word vector corresponding to each target data through a preset word embedding model; wherein, the preset word embedding model is trained by sample data and the sample word vectors corresponding to the sample data;

[0024] Perform weighted processing on each initial word vector according to the sentiment score corresponding to each evaluation data to obtain a target word vector;

[0025] Input the target word vector into a preset sentiment analysis model to obtain the sentiment tendency corresponding to each target word vector; wherein, the preset sentiment analysis model is trained by sample word vectors and the sample sentiment tendencies corresponding to the sample word vectors.

[0026] In yet another alternative of the first aspect, preprocessing each evaluation data to obtain target data includes:

[0027] Performing segmentation processing on each evaluation data to obtain at least two phrase data;

[0028] Performing deletion processing on each phrase data based on a preset character set to obtain refined data;

[0029] Performing regularization processing on all refined data to obtain target data.

[0030] In yet another alternative of the first aspect, based on the sentiment tendency and statistical data, generating an analysis result corresponding to the gas station, including:

[0031] Obtaining improvement dimension words according to the sentiment tendency and statistical data; wherein, the preset dimension words include improvement dimension words;

[0032] Obtaining detection data corresponding to the improvement dimension words in the gas station, and generating corresponding improvement suggestions according to the detection data;

[0033] Performing visualization processing on the improvement dimension words, detection data, and improvement suggestions to obtain an analysis result corresponding to the gas station.

[0034] In a second aspect, an embodiment of the present application provides a sentiment analysis device based on gas station customer evaluations, including:

[0035] A first processing module, configured to establish a sentiment dictionary based on at least two evaluation data corresponding to the gas station, and determine a sentiment score corresponding to each evaluation data according to the sentiment dictionary; wherein, the sentiment dictionary includes at least two types of sentiment texts generated from the evaluation data and sentiment scores corresponding to each type of sentiment text;

[0036] A second processing module, configured to determine a sentiment tendency corresponding to each evaluation data based on each evaluation data and the corresponding sentiment score;

[0037] A third processing module, configured to perform classification processing on all evaluation data based on the sentiment tendency to obtain at least three types of labeled data, and count the number of occurrences of at least two preset dimension words in each type of labeled data to obtain statistical data;

[0038] A fourth processing module, configured to generate an analysis result corresponding to the gas station based on the sentiment tendency and statistical data.

[0039] In a third aspect, an embodiment of the present application further provides a sentiment analysis device based on gas station customer evaluations, including a processor and a memory;

[0040] The processor is connected to the memory;

[0041] A memory for storing executable program code;

[0042] The processor runs a program corresponding to the executable program code by reading the executable program code stored in the memory, so as to implement the sentiment analysis method based on gas station customer evaluations provided in the first aspect or any implementation manner of the first aspect of the embodiments of the present application.

[0043] In a fourth aspect, the embodiments of the present application provide a computer storage medium. The computer storage medium stores a computer program, and the computer program includes program instructions. When the program instructions are executed by a processor, the sentiment analysis method based on gas station customer evaluations provided in the first aspect or any implementation manner of the first aspect of the embodiments of the present application can be implemented.

[0044] The beneficial effects brought by the technical solutions provided in some embodiments of this specification at least include:

[0045] In the process of sentiment analysis based on gas station customer evaluations, a sentiment dictionary is established according to the customer evaluation data of the gas station, and then the sentiment tendency in the customer evaluation data is accurately identified through this dictionary, and the number of occurrences of dimension words related to the gas station in each sentiment tendency in the customer evaluation data is counted. Thus, the analysis result is obtained according to the sentiment tendency recognition result and the statistical result, improving the accuracy of sentiment analysis of customer evaluation data, realizing the effective utilization of customer feedback, being conducive to better improving the service quality of gas stations, and thus more effectively enhancing customer satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0047] Figure 1 It is the overall flowchart of a sentiment analysis method based on gas station customer evaluations provided by the embodiments of the present application;

[0048] Figure 2 It is the overall flowchart of another sentiment analysis method based on gas station customer evaluations provided by the embodiments of the present application;

[0049] Figure 3 It is the structural schematic diagram of a sentiment analysis device based on gas station customer evaluations provided by the embodiments of the present application;

[0050] Figure 4It is a schematic structural diagram of another emotional analysis device based on gas station customer evaluations provided by the embodiments of the present application. Detailed implementation manners

[0051] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application.

[0052] In the following introduction, the terms "first" and "second" are only for the purpose of description and cannot be construed as indicating or implying relative importance. The following introduction provides multiple embodiments of the present application. Different embodiments can be replaced or combined, so the present application can also be considered to include all possible combinations of the same and / or different embodiments described. Thus, if one embodiment includes features A, B, and C, and another embodiment includes features B and D, then the present application should also be considered to include embodiments containing one or more of all other possible combinations of A, B, C, and D, even though such embodiments may not be explicitly described in the following content.

[0053] The following description provides examples and does not limit the scope, applicability, or examples set forth in the claims. Changes can be made to the functions and arrangements of the described elements without departing from the scope of the content of the present application. Each example can appropriately omit, substitute, or add various processes or components. For example, the described methods can be executed in a different order than the described order, and various steps can be added, omitted, or combined. In addition, the features described in some examples can be combined into other examples.

[0054] Please refer to Figure 1 , Figure 1 which shows an overall flowchart of an emotional analysis method based on gas station customer evaluations provided by the embodiments of the present application.

[0055] As Figure 1 shown, the emotional analysis method based on gas station customer evaluations can at least include the following steps:

[0056] Step 101: Establish an emotional dictionary based on at least two evaluation data corresponding to the gas station, and determine an emotional score corresponding to each evaluation data according to the emotional dictionary.

[0057] In the embodiments of the present application, the sentiment analysis method based on gas station customer evaluations can be, but is not limited to, applied on mobile terminals, servers, or cloud platforms. During the sentiment analysis process based on gas station customer evaluations, a sentiment dictionary is established according to the customer evaluation data of the gas station, and then the sentiment tendency in the customer evaluation data is accurately identified through this dictionary. The number of occurrences of dimension words related to the gas station in each sentiment tendency in the customer evaluation data is counted. Thus, the analysis result is obtained based on the sentiment tendency identification result and the statistical result, improving the accuracy of the sentiment analysis of customer evaluation data, realizing the effective utilization of customer feedback, facilitating better improvement of the service quality of the gas station, and thus more effectively enhancing customer satisfaction.

[0058] Specifically, during the sentiment analysis process based on gas station customer evaluations, a sentiment dictionary can be first established according to at least two evaluation data corresponding to the gas station. Among them, the sentiment dictionary includes at least two sentiment texts generated from the evaluation data and the sentiment scores corresponding to each sentiment text.

[0059] It can be understood that the ways to obtain customer evaluation data can be, but are not limited to, directly accessing the evaluation data through the internal database of the company to which the gas station belongs, collecting evaluation data from the mini-program corresponding to the gas station using web crawler technology, and conducting offline questionnaires on customers. During the process of establishing the sentiment dictionary, the Chinese characters or words with emotional colors and gas station-related terms in the evaluation data can be first identified through regular expressions to obtain the sentiment texts.

[0060] It should be noted that regular expressions (Regular Expression) are usually abbreviated as regex or regexp. It is a powerful text processing tool commonly used for operations such as pattern matching, searching, and replacement. It can match a class of strings through a "template" created by the user, rather than just matching an exact string. Here, a list containing Chinese characters or words with emotional colors can be constructed as the matching template. This list can be created manually or obtained from existing sentiment dictionaries, such as the Chinese version of NRC Emotion Lexicon or other sentiment dictionaries specifically for Chinese. Then, the texts in the evaluation data containing the sentiment words in the list can be identified through regular expressions. Suppose we have a sentiment word list: "good", "friendly", "bad", "dissatisfied", "reasonable", "not bad", and there are two evaluation data: "The service attitude of the staff is friendly" and "The oil price is reasonable". Then, the content that can be identified through regular expressions is "friendly" and "reasonable". Similarly, gas station-related terms such as "service attitude" and "oil price" can be identified through regular expressions, and then the sentiment texts "The service attitude is friendly" and "The oil price is reasonable" can be obtained.

[0061] After obtaining the sentiment texts, one can, but is not limited to, assign corresponding sentiment scores to each sentiment text by means such as expert annotation, referring to the set scores of other existing sentiment dictionaries, or custom rules (such as setting according to the frequency of the appearance of the sentiment text), and merge the sentiment texts with the same content as one kind of sentiment text, so as to establish a sentiment dictionary specifically for gas stations.

[0062] Furthermore, after establishing the sentiment dictionary, the sentiment scores of all sentiment texts in each evaluation data can be determined through the sentiment dictionary, and then a sentiment score corresponding to the evaluation data can be calculated based on these sentiment scores. For example, when there is an evaluation data "The service attitude of this gas station is good and the oil price is reasonable", which contains two sentiment texts "good service attitude" and "reasonable oil price", if the sentiment scores corresponding to these two sentiment texts are found to be 1.5 and 1.0 respectively through the established sentiment dictionary, then one can, but is not limited to, calculate the sentiment score corresponding to the evaluation data by means such as taking the sum, taking the average value, taking the weighted average value, taking the median, taking the maximum (or minimum) value, or other value-taking methods based on these two sentiment scores, such as 1.5 + 1.0 = 2.5 or (1.5 + 1.0) / 2 = 1.25, etc.

[0063] As an option of the embodiment of the present application, establishing a sentiment dictionary based on at least two evaluation data corresponding to a gas station includes:

[0064] Extract at least two sentiment texts from at least two evaluation data corresponding to the gas station;

[0065] Perform annotation processing on all sentiment texts, and calculate the sentiment score corresponding to each sentiment text according to the result of the annotation processing;

[0066] Establish a sentiment dictionary based on all sentiment texts and the sentiment scores corresponding to each sentiment text.

[0067] Specifically, in the process of establishing an emotion dictionary based on evaluation data, corresponding emotion texts can be extracted from the evaluation data first. The extraction methods can include but are not limited to rule matching (such as regular expressions), topic modeling (such as LDA or NMF), etc. After the emotion texts are extracted, annotation processing can be performed on the emotion texts. The annotation methods can include but are not limited to annotating emotion polarity and emotion intensity, that is, judging the emotion polarity (such as "positive", "negative", "neutral", etc.) and emotion intensity (such as "strong", "relatively strong", "medium", "relatively weak", "weak", etc.) of each emotion text and organizing them into a data structure or a relational database, and then obtaining the result of the annotation processing (such as "good service attitude" → "positive" and "relatively strong"), so that different calculation methods can be adopted for emotion texts with different polarities according to this result to obtain the emotion score corresponding to each emotion text. For example, specific scores can be set corresponding to the emotion polarity and emotion intensity as the emotion scores. If the emotion intensity is divided into the above five intensities, the emotion scores can be set for the emotion texts with the emotion polarity of "positive" as: 1, 3, 5, 7, 9 (the larger the score, the stronger the emotion intensity). At the same time, the emotion scores can be set for the emotion texts with the emotion polarity of "negative" as: -2, -4, -6, -8, -10 (the smaller the score, the stronger the emotion intensity), and the emotion score for the emotion texts with the emotion polarity of "neutral" is set to 0.

[0068] It should be noted that the annotation processing can be realized through but not limited to manual annotation, automated or semi-automated tools and platforms (such as Brat, Label Studio, Doccano, etc.).

[0069] Next, after obtaining the emotion scores of all emotion texts, the emotion texts with the same content can be merged, and then a corresponding data structure or database can be established as the emotion dictionary.

[0070] As another option of the embodiment of the present application, performing annotation processing on all emotion texts and calculating the emotion score corresponding to each emotion text according to the result of the annotation processing includes:

[0071] Performing annotation processing on all emotion texts and determining the numerical range of the emotion score corresponding to each emotion text according to the result of the annotation processing;

[0072] Counting the number of the same emotion texts to obtain a first value;

[0073] Counting the number of all emotion texts to obtain a second value;

[0074] Calculating the emotion score corresponding to each emotion text according to the first value, the second value and the numerical range.

[0075] Specifically, in the process of annotating emotional texts and calculating the emotional scores corresponding to all emotional texts based on the results of the annotation process, the emotional polarity of all emotional texts can be marked first, so as to determine the range of emotional scores corresponding to each emotional text. For example, emotional texts with a positive emotional polarity are assigned an emotional score range of 0 to 5 (including 1 but not including 0), emotional texts with a negative emotional polarity are assigned an emotional score range of -5 to 0 (including -5 but not including 0), and emotional texts with a neutral emotional polarity are assigned an emotional score of 0.

[0076] Next, the number of the same emotional texts (i.e., emotional texts with the same content) and the number of all emotional texts can be counted to obtain the corresponding first value and second value. For example, when there are three emotional texts: "good service attitude", "reasonable price", "good service attitude", for the emotional text "good service attitude", the first value is 2 and the second value is 3; for the emotional text "reasonable price", the first value is 1 and the second value is 3.

[0077] After that, since the emotional polarities of these two emotional texts are positive and neutral respectively (assuming that the emotional polarity of the emotional text expressing affirmation of the gas station is positive), referring to the example of the emotional score range division in the previous paragraph, the emotional score corresponding to "good service attitude" can be calculated based on "2", "3" and "0 to 5", and the emotional score corresponding to "reasonable price" can also be obtained as 0 according to "0".

[0078] It should be noted that the ways to calculate the specific emotional scores based on the first value, the second value and the numerical range of the emotional scores can be, but are not limited to, the following two:

[0079] (1) First, calculate the ratio x of the first value to the second value (such as 2 / 3) and the length y of the numerical range (such as 5), and then multiply x by y, and the product is used as the specific emotional score;

[0080] (2) When the first value is greater than half of the second value, the first value is used as the specific emotional score; when the first value is less than or equal to half of the second value, the product of the median value of the numerical range (such as 2.5) and the first value is used as the specific emotional score.

[0081] As another option of the embodiment of the present application, calculating the emotional score corresponding to each emotional text according to the first value, the second value and the numerical range includes:

[0082] Determining an intensity coefficient according to the ratio of the first value to the second value;

[0083] Calculate the sentiment score corresponding to each sentiment text according to the intensity coefficient and the numerical range.

[0084] Specifically, after obtaining the first value, the second value, and the corresponding numerical range, the ratio of the first value to the second value can be calculated first as the intensity coefficient, which is used to represent the degree of importance that the customer attaches to the aspect represented by the sentiment text. For example, when the number of sentiment texts such as "poor environment" is large, it means that the customer cares more about the environment aspect of the gas station.

[0085] Then, the intensity coefficient can be multiplied by the length of the numerical range, and the product obtained is the sentiment score corresponding to the sentiment text.

[0086] Step 102: Based on each evaluation data and the corresponding sentiment score, determine the sentiment tendency corresponding to each evaluation data.

[0087] Specifically, after determining each evaluation data and the corresponding sentiment score, the sentiment tendency of each evaluation data can be determined according to the value of the sentiment score and the preset range. For example, there is an evaluation data A with a sentiment score of 36, an evaluation data B with a sentiment score of -20, and an evaluation data C with a sentiment score of 8. The preset ranges are X: -60 to -10 (including -60 but not including -10), Y: -10 to 10 (including -10 and including 10), Z: 10 to 60 (including 60 but not including 10). Then, the determination rule of the sentiment tendency can be set as follows: when the sentiment score of the evaluation data is within the X range, it is determined that the sentiment tendency of the evaluation data is negative; when the sentiment score of the evaluation data is within the Y range, it is determined that the sentiment tendency of the evaluation data is neutral; when the sentiment score of the evaluation data is within the Z range, it is determined that the sentiment tendency of the evaluation data is positive; when the sentiment score of the evaluation data is not within the preset range, the evaluation data is invalidated (i.e., deleted from the storage space).

[0088] As another alternative of the embodiment of the present application, determining the sentiment tendency corresponding to each evaluation data based on each evaluation data and the corresponding sentiment score includes:

[0089] Preprocess each evaluation data to obtain target data, and generate an initial word vector corresponding to each target data through a preset word embedding model; wherein, the preset word embedding model is trained by sample data and sample word vectors corresponding to the sample data;

[0090] According to the sentiment score corresponding to each evaluation data, perform weighted processing on each initial word vector to obtain a target word vector;

[0091] Input the target word vectors into a preset sentiment analysis model to obtain the sentiment tendency corresponding to each target word vector; wherein, the preset sentiment analysis model is trained by sample word vectors and the sample sentiment tendencies corresponding to the sample word vectors.

[0092] Specifically, in the process of determining the sentiment tendency corresponding to each evaluation data, each evaluation data can be preprocessed first to obtain target data (such as text data of string type, etc.) suitable for input into a preset word embedding model (such as Word2Vec model, GloVe model, FastText model, BERT model, ELMo model, etc.), and the initial word vectors corresponding to the target data are generated through this model. It should be noted that the preset word embedding model is trained by sample data and the sample word vectors corresponding to the sample data.

[0093] Then, the sentiment scores corresponding to each evaluation data can be used as weights, and the initial word vectors can be weighted processed through weighted algorithms such as TF-IDF, SIF, etc., but not limited to these, and then the target word vectors can be obtained.

[0094] After that, the target word vectors can be input into a preset sentiment analysis model to obtain the sentiment tendency corresponding to each target word vector. It should be noted that the preset sentiment analysis model is trained by sample word vectors and the sample sentiment tendencies corresponding to the sample word vectors, and moreover, this model is realized based on the fusion of the MacBERT model and the TextCNN model. It can be understood that the target word vectors are input into the MacBERT model to obtain the corresponding output results, and then the output results are used as the input data of the TextCNN model, so as to obtain the sentiment tendency.

[0095] As another option in the embodiments of the present application, preprocessing each evaluation data to obtain target data includes:

[0096] Performing segmentation processing on each evaluation data to obtain at least two phrase data;

[0097] Performing deletion processing on each phrase data based on a preset character set to obtain refined data;

[0098] Performing regularization processing on all the refined data to obtain target data.

[0099] Specifically, in the process of preprocessing the evaluation data to obtain the target data, the Jieba (a Chinese word segmentation library) word segmentation technology can be used to segment each evaluation data first, obtaining at least two phrase data. After that, a predefined stop word list can be loaded, which contains common but meaningless words such as "de", "le", "zai", etc. Then, Python scripts can be used to loop through and compare to remove these stop words, so as to achieve the deletion process for each phrase data, thereby obtaining the streamlined data and ensuring that only meaningful words are retained in the target data.

[0100] Next, all the streamlined data can be regularized, but not limited to, through regular expressions to obtain the standardized target data. Among them, the regularization process can make the data consistent in case, expand common abbreviations into full forms or correct spelling mistakes, enabling the model to better understand and process the data.

[0101] Step 103: Classify all the evaluation data based on the sentiment tendency to obtain at least three categories of labeled data, and count the number of occurrences of at least two predefined dimension words in each category of labeled data to obtain the statistical data.

[0102] Specifically, after determining the sentiment tendency of each evaluation data, all the evaluation data can be classified according to the sentiment tendency (i.e., positive, negative, neutral or good review, bad review, medium review, etc.) to obtain at least three categories of labeled data. Among them, the evaluation data in each category of labeled data have the same sentiment tendency.

[0103] Next, the number of occurrences of at least two predefined dimension words in each category of labeled data can be counted, and then the statistical data can be obtained. Among them, the predefined dimension words mainly include evaluation terms related to gas stations, such as "environment", "service", "oil price", etc. It can be understood that the statistical data at least includes content such as: "negative → service → 30", "positive → facilities → 150" or "the slow service speed is mentioned 30 times in bad reviews", "the fueling facilities are perfect and are mentioned 150 times in good reviews", etc.

[0104] Step 104: Generate the analysis result corresponding to the gas station based on the sentiment tendency and the statistical data.

[0105] Specifically, after obtaining the sentiment tendency and statistical data of the evaluation data, based on the sentiment tendency and statistical data of all the evaluation data of the gas station, the aspects that need to be improved and feasible measures of the gas station can be analyzed. For example, if the proportion of data with a negative sentiment tendency in the evaluation data is relatively large, it indicates that there are many aspects that need to be improved in the gas station. The specific aspects that need to be improved can be determined according to the dimensional words that appear more frequently in the categories with a negative sentiment tendency in the statistical data. In addition, feasible measures can be determined based on the direction of improvement, the collection source of the evaluation data, the collection time, etc. For example, when "poor service attitude" is mentioned the most times in the negative reviews and the evaluation data containing this content is concentrated in a certain period of a certain month, the staff list for that month and that period can be queried, and they can be trained and educated on service attitude.

[0106] As another option of the embodiment of the present application, based on the sentiment tendency and statistical data, an analysis result corresponding to the gas station is generated, including:

[0107] According to the sentiment tendency and statistical data, improvement dimensional words are obtained; among them, the preset dimensional words include improvement dimensional words;

[0108] The detection data corresponding to the improvement dimensional words in the gas station is obtained, and corresponding improvement suggestions are generated according to the detection data;

[0109] The improvement dimensional words, the detection data, and the improvement suggestions are visually processed to obtain an analysis result corresponding to the gas station.

[0110] Specifically, in the process of generating an analysis result corresponding to the gas station according to the sentiment tendency and statistical data of all the evaluation data of the gas station, the improvement dimensional words can be first determined according to the sentiment tendency and statistical data, that is, the dimensional words with a negative sentiment tendency or negative reviews in the statistical data (it can be the dimensional word with the most occurrences, or multiple dimensional words with more occurrences).

[0111] Then, the detection data corresponding to the improvement dimensional words in the gas station can be obtained. For example, when the improvement dimensional words include "oil price" and "service", data such as the recent oil product supply list and the staff list of the gas station can be obtained. Furthermore, the specific deficiencies of the gas station can be found according to these data, and thus a specific adjustment plan (that is, an improvement suggestion, such as changing the supplier, adjusting the working hours of the staff, replacing with new and efficient equipment, etc.) can be formulated for the deficiencies to improve the industry competitiveness and enhance user satisfaction.

[0112] After that, the improved dimension words, detection data, and improvement suggestions can also be visualized. Among them, the visualization process can be but is not limited to being implemented through tools such as Tableau, Power BI, Qlik Sense / QlikView, or libraries such as Grafana, Matplotlib, Seaborn, Plotly, D3.js, etc., and then the analysis results corresponding to the gas station can be obtained. It can be understood that the analysis results at least include the important aspects that need to be improved (i.e., the improved dimension words), the specific deficiencies shown by the detection data, and the measures that can be taken for the deficiencies, etc.

[0113] Please refer to Figure 2 , Figure 2 which shows the overall flowchart of another sentiment analysis method based on gas station customer evaluations provided by an embodiment of the present application.

[0114] As Figure 2 shown, in the process of sentiment analysis based on gas station customer evaluations, the evaluation data of customers can be collected first through multiple platforms and websites of the gas station, including the overall evaluation of the gas station by customers and the feedback on specific services; then, a sentiment dictionary specifically for the gas station can be established according to these evaluation data to distinguish different sentiment polarities and different sentiment intensities expressed by the evaluation data; then, the collected evaluation data can be preprocessed, including text cleaning, removing irrelevant information, and language standardization, etc., to obtain the target data; thus, sentiment analysis processing can be performed on the target data according to the sentiment dictionary and the preset sentiment analysis model to obtain the sentiment tendency of each evaluation data; then, the number of times each dimension word appears in each sentiment tendency can be counted to obtain the statistical data to determine the aspects that need to be improved for the gas station; after that, analysis data (such as the above-mentioned improved dimension words, improvement suggestions, etc.) can be generated according to the statistical data and the sentiment tendency of each evaluation data, and the analysis data can be visualized to obtain the analysis results; after obtaining the analysis results, the analysis results can be integrated into the website or management system of the gas station through API (Application Programming Interface, which is a set of predefined functions, protocols, and tools, and its role is to allow different software programs to communicate and exchange data without understanding the underlying implementation details) interface technology, so that managers can conveniently access and view these data to achieve timely feedback and improvement. At the same time, a customer interaction function can also be provided through the website to allow customers to make more informed choices and feedback based on the visualized analysis results, thereby promoting the continuous improvement and optimization of the site.

[0115] Please refer to Figure 3 , Figure 3The figure shows a schematic structural diagram of an emotion analysis device based on gas station customer evaluations provided by an embodiment of the present application.

[0116] As Figure 3 shown, the emotion analysis device based on gas station customer evaluations may at least include a first processing module 301, a second processing module 302, a third processing module 303, and a fourth processing module 304, where:

[0117] The first processing module 301 is configured to establish an emotion dictionary based on at least two evaluation data corresponding to the gas station, and determine an emotion score corresponding to each evaluation data according to the emotion dictionary; wherein, the emotion dictionary includes at least two emotion texts generated from the evaluation data and an emotion score corresponding to each emotion text;

[0118] The second processing module 302 is configured to determine an emotion tendency corresponding to each evaluation data based on each evaluation data and the corresponding emotion score;

[0119] The third processing module 303 is configured to classify all the evaluation data based on the emotion tendency to obtain at least three types of labeled data, and count the number of occurrences of at least two preset dimensional words in each type of labeled data to obtain statistical data;

[0120] The fourth processing module 304 is configured to generate an analysis result corresponding to the gas station based on the emotion tendency and the statistical data.

[0121] In some possible embodiments, establishing an emotion dictionary based on at least two evaluation data corresponding to the gas station includes:

[0122] Specifically, the first processing module 301 is configured to:

[0123] Extract at least two emotion texts from at least two evaluation data corresponding to the gas station;

[0124] Perform annotation processing on all the emotion texts, and calculate an emotion score corresponding to each emotion text according to the result of the annotation processing;

[0125] Establish an emotion dictionary based on all the emotion texts and the emotion score corresponding to each emotion text.

[0126] In some possible embodiments, performing annotation processing on all the emotion texts and calculating an emotion score corresponding to each emotion text according to the result of the annotation processing includes:

[0127] Specifically, the first processing module 301 is configured to:

[0128] Perform annotation processing on all the emotion texts, and determine the numerical range of the emotion score corresponding to each emotion text according to the result of the annotation processing;

[0129] Count the number of texts with the same sentiment to obtain a first value;

[0130] Count the number of all sentiment texts to obtain a second value;

[0131] Calculate the sentiment score corresponding to each sentiment text according to the first value, the second value, and the value range.

[0132] In some possible embodiments, calculating the sentiment score corresponding to each sentiment text according to the first value, the second value, and the value range includes:

[0133] The first processing module 301 is specifically configured to:

[0134] Determine an intensity coefficient according to the ratio of the first value to the second value;

[0135] Calculate the sentiment score corresponding to each sentiment text according to the intensity coefficient and the value range.

[0136] In some possible embodiments, determining the sentiment tendency corresponding to each evaluation data based on each evaluation data and the corresponding sentiment score includes:

[0137] The second processing module 302 is specifically configured to:

[0138] Preprocess each evaluation data to obtain target data, and generate an initial word vector corresponding to each target data through a preset word embedding model; wherein, the preset word embedding model is trained by sample data and sample word vectors corresponding to the sample data;

[0139] Perform weighted processing on each initial word vector according to the sentiment score corresponding to each evaluation data to obtain a target word vector;

[0140] Input the target word vector into a preset sentiment analysis model to obtain the sentiment tendency corresponding to each target word vector; wherein, the preset sentiment analysis model is trained by sample word vectors and sample sentiment tendencies corresponding to the sample word vectors.

[0141] In some possible embodiments, preprocessing each evaluation data to obtain target data includes:

[0142] The second processing module 302 is specifically configured to:

[0143] Perform segmentation processing on each evaluation data to obtain at least two phrase data;

[0144] Perform deletion processing on each phrase data based on a preset character set to obtain refined data;

[0145] Regularize all the refined data to obtain the target data.

[0146] In some possible embodiments, based on the sentiment tendency and statistical data, generate an analysis result corresponding to the gas station, including:

[0147] The fourth processing module 304 is specifically configured to:

[0148] Obtain improvement dimension words according to the sentiment tendency and statistical data; wherein, the preset dimension words include improvement dimension words;

[0149] Obtain the detection data corresponding to the improvement dimension words in the gas station, and generate corresponding improvement suggestions according to the detection data;

[0150] Perform visualization processing on the improvement dimension words, detection data, and improvement suggestions to obtain an analysis result corresponding to the gas station.

[0151] Please refer to Figure 4 , Figure 4 which shows a schematic structural diagram of another sentiment analysis device based on gas station customer evaluations provided by an embodiment of the present application.

[0152] As Figure 4 shown, the sentiment analysis device 400 based on gas station customer evaluations may include at least one processor 401, at least one network interface 404, a user interface 403, a memory 405, and at least one communication bus 402.

[0153] Among them, the communication bus 402 can be used to realize the connection and communication of the above-mentioned various components.

[0154] Among them, the user interface 403 may include buttons, and the optional user interface may further include a standard wired interface and a wireless interface.

[0155] Among them, the network interface 404 may but is not limited to include a Bluetooth module, an NFC module, a Wi-Fi module, etc.

[0156] Among them, the processor 401 may include one or more processing cores. The processor 401 connects various parts within the entire sentiment analysis device 400 based on gas station customer evaluations through various interfaces and lines. By running or executing instructions, programs, code sets, or instruction sets stored in the memory 405, and by calling the data stored in the memory 405, it executes various functions of the sentiment analysis device 400 based on gas station customer evaluations and processes data. Optionally, the processor 401 may be implemented in at least one hardware form of DSP, FPGA, or PLA. The processor 401 may integrate one or a combination of several of CPU, GPU, and modem, etc. Among them, the CPU mainly processes the operating system, user interface, application programs, etc.; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 401 and may be implemented separately by a single chip.

[0157] Among them, the memory 405 may include RAM and may also include ROM. Optionally, the memory 405 includes a non-transitory computer-readable medium. The memory 405 can be used to store instructions, programs, code, code sets, or instruction sets. The memory 405 may include a program storage area and a data storage area. Among them, the program storage area may store instructions for implementing the operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store the data involved in the above-mentioned various method embodiments. Optionally, the memory 405 may also be at least one storage device located far from the aforementioned processor 401. As Figure 4 shown, the memory 405 as a computer storage medium may include an operating system, a network communication module, a user interface module, and a sentiment analysis application program based on gas station customer evaluations.

[0158] Specifically, the processor 401 may be used to call the sentiment analysis application program stored in the memory 405 and specifically perform the following operations:

[0159] Establish a sentiment dictionary based on at least two evaluation data corresponding to a gas station, and determine a sentiment score corresponding to each evaluation data according to the sentiment dictionary; wherein, the sentiment dictionary includes at least two sentiment texts generated from the evaluation data and the sentiment scores corresponding to each sentiment text;

[0160] Based on each evaluation data and the corresponding sentiment score, determine the sentiment tendency corresponding to each evaluation data;

[0161] Classify all evaluation data based on sentiment tendency to obtain at least three categories of labeled data, and count the number of occurrences of at least two preset dimensional words in each category of labeled data to obtain statistical data;

[0162] Generate an analysis result corresponding to the gas station based on the sentiment tendency and the statistical data.

[0163] This application also provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, the steps of the above method are implemented. Among them, the computer-readable storage medium may include, but is not limited to, any type of disk, including floppy disks, optical disks, DVDs, CD-ROMs, microdrives, and magneto-optical disks, ROMs, RAMs, EPROMs, EEPROMs, DRAMs, VRAMs, flash memory devices, magnetic cards or optical cards, nanosystems (including molecular memory ICs), or any type of medium or device suitable for storing instructions and / or data.

[0164] It should be noted that for the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that this application is not limited by the described action sequence, because according to this application, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0165] In the above embodiments, the descriptions of the various embodiments have their own emphases. For the parts not detailed in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0166] In several embodiments provided by this application, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point, the displayed or discussed mutual coupling or direct coupling or communication connection can be through some service interfaces. The indirect coupling or communication connection of the device or unit can be in an electrical or other form.

[0167] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place, or they can be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0168] In addition, in each embodiment of the present application, each functional unit can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

[0169] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in each embodiment of the present application. The aforementioned memory includes: various media such as USB flash drives, read-only memories (ROM), random access memories (RAM), external hard drives, magnetic disks, or optical discs that can store program codes.

[0170] Those of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing relevant hardware through a program. This program can be stored in a computer-readable memory, and the memory can include: flash drives, read-only memories (ROM), random access memories (RAM), magnetic disks, or optical discs, etc.

[0171] The above are only exemplary embodiments of the present disclosure, and the scope of the present disclosure cannot be limited thereby. That is, any equivalent changes and modifications made in accordance with the teachings of the present disclosure still fall within the scope covered by the present disclosure. Those skilled in the art will easily think of other implementation manners of the present disclosure after considering the specification and practicing the present disclosure herein. The present application aims to cover any variations, uses, or adaptive changes of the present disclosure. These variations, uses, or adaptive changes follow the general principles of the present disclosure and include common general knowledge or conventional technical means in the technical field not recorded in the present disclosure. The specification and the embodiments are only regarded as exemplary, and the scope and spirit of the present disclosure are defined by the claims.

Claims

1. A sentiment analysis method based on gas station customer reviews, characterized in that: include: Establishing a sentiment dictionary based on at least two evaluation data corresponding to the gas station, and determining a sentiment score corresponding to each evaluation data according to the sentiment dictionary; wherein the sentiment dictionary includes at least two sentiment texts generated by the evaluation data and a sentiment score corresponding to each sentiment text; Based on each of the evaluation data and the corresponding sentiment score, determining the sentiment tendency corresponding to each of the evaluation data; Classify all the evaluation data based on the sentiment tendency to obtain at least three categories of label data, and count the number of occurrences of at least two preset dimension words in each category of the label data to obtain statistical data; Based on the sentiment tendency and the statistical data, generating an analysis result corresponding to the gas station; The step of establishing a sentiment dictionary based on at least two evaluation data corresponding to the gas station includes: Extracting at least two sentiment texts from at least two evaluation data corresponding to the gas station; Performing annotation processing on all the emotion texts, and calculating the emotion score corresponding to each emotion text according to the result of the annotation processing; Establishing a sentiment dictionary based on all the sentiment texts and the sentiment score corresponding to each of the sentiment texts; The tagging of all the emotion texts and calculating the emotion score corresponding to each emotion text according to the tagging result include: Performing a tagging process on all the emotion texts, and determining a numerical range of an emotion score corresponding to each emotion text according to the result of the tagging process; Counting the number of the same sentiment texts to obtain a first value; Counting the number of all the sentiment texts to obtain a second value; A sentiment score corresponding to each of the sentiment texts is calculated according to the first value, the second value, and the value range.

2. The method according to claim 1, characterized in that The calculating, according to the first value, the second value and the value range, a sentiment score corresponding to each of the sentiment texts comprises: Determining an intensity coefficient according to a ratio of the first value to the second value; The sentiment score corresponding to each of the sentiment texts is calculated according to the intensity coefficient and the numerical range.

3. The method according to claim 1, characterized in that The step of determining the emotional tendency corresponding to each evaluation data based on each evaluation data and the corresponding emotional score includes: Preprocessing each of the evaluation data to obtain target data, and generating an initial word vector corresponding to each of the target data through a preset word embedding model; wherein the preset word embedding model is trained by sample data and sample word vectors corresponding to the sample data; Performing weighted processing on each of the initial word vectors according to the sentiment score corresponding to each of the evaluation data to obtain a target word vector; The target word vector is input into a preset sentiment analysis model to obtain the sentiment tendency corresponding to each target word vector; wherein the preset sentiment analysis model is trained by sample word vectors and sample sentiment tendencies corresponding to the sample word vectors.

4. The method according to claim 3, characterized in that The preprocessing of each evaluation data to obtain target data comprises: Segmenting each evaluation data to obtain at least two phrase data; Performing deletion processing on each of the phrase data based on a preset character set to obtain simplified data; Regularization processing is performed on all the simplified data to obtain target data.

5. The method according to claim 1, characterized in that: The generating of the analysis result corresponding to the gas station based on the sentiment tendency and the statistical data includes: According to the sentiment tendency and the statistical data, an improved dimensional word is obtained; wherein the preset dimensional word includes the improved dimensional word; Acquire detection data corresponding to the improvement dimension word in the gas station, and generate corresponding improvement suggestions according to the detection data; The improvement dimension words, the detection data and the improvement suggestions are visualized to obtain analysis results corresponding to the gas station.

6. A sentiment analysis device based on gas station customer reviews, characterized in that: include: A first processing module is used to establish a sentiment dictionary based on at least two evaluation data corresponding to the gas station, and determine a sentiment score corresponding to each of the evaluation data according to the sentiment dictionary; wherein the sentiment dictionary includes at least two sentiment texts generated by the evaluation data and a sentiment score corresponding to each of the sentiment texts; A second processing module, configured to determine the sentiment tendency corresponding to each evaluation data based on each evaluation data and the corresponding sentiment score; A third processing module is used to classify all the evaluation data based on the sentiment tendency to obtain at least three types of label data, and count the number of occurrences of at least two preset dimension words in each type of the label data to obtain statistical data; a fourth processing module, configured to generate an analysis result corresponding to the gas station based on the sentiment tendency and the statistical data; A sentiment dictionary is established based on at least two evaluation data corresponding to the gas station, including: The first processing module is specifically used for: Extracting at least two sentiment texts from at least two evaluation data corresponding to the gas station; All sentiment texts are annotated, and the sentiment score corresponding to each sentiment text is calculated based on the annotated results; Building a sentiment dictionary based on all sentiment texts and the sentiment scores corresponding to each sentiment text; All sentiment texts are annotated, and the sentiment scores corresponding to each sentiment text are calculated based on the annotated results, including: The first processing module is also specifically used for: All sentiment texts are annotated, and the numerical range of the sentiment score corresponding to each sentiment text is determined according to the result of the annotation process; Count the number of texts with the same sentiment to obtain a first value; Count the number of all sentiment texts to obtain a second value; A sentiment score corresponding to each sentiment text is calculated according to the first value, the second value, and the value range.

7. A sentiment analysis device based on gas station customer reviews, characterized in that: including a processor and a memory; The processor is connected to the memory; The memory is used to store executable program code; The processor runs a program corresponding to the executable program code by reading the executable program code stored in the memory, so as to execute the steps of the method according to any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed on a computer or a processor, the computer or the processor executes the steps of the method according to any one of claims 1 to 5.

Images