WO2023159766A1 - 餐饮数据分析方法、装置、电子设备及存储介质 - Google Patents

餐饮数据分析方法、装置、电子设备及存储介质 Download PDF

Info

Publication number
WO2023159766A1
WO2023159766A1 PCT/CN2022/090739 CN2022090739W WO2023159766A1 WO 2023159766 A1 WO2023159766 A1 WO 2023159766A1 CN 2022090739 W CN2022090739 W CN 2022090739W WO 2023159766 A1 WO2023159766 A1 WO 2023159766A1
Authority
WO
WIPO (PCT)
Prior art keywords
visualization
analysis
catering
data
words
Prior art date
Application number
PCT/CN2022/090739
Other languages
English (en)
French (fr)
Inventor
刘羲
舒畅
陈又新
Original Assignee
平安科技（深圳）有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 平安科技（深圳）有限公司 filed Critical 平安科技（深圳）有限公司
Publication of WO2023159766A1 publication Critical patent/WO2023159766A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/34—Browsing; Visualisation therefor
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/20—Natural language analysis
    • G06F40/253—Grammatical analysis; Style critique
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/20—Natural language analysis
    • G06F40/279—Recognition of textual entities
    • G06F40/284—Lexical analysis, e.g. tokenisation or collocates
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/30—Semantic analysis
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management

Definitions

  • the present application relates to the technical fields of computer, artificial intelligence and natural language processing, and in particular to a method, device, electronic equipment and storage medium for analyzing catering data.
  • the online market has also become a new rapid growth point for enterprises.
  • the Internet Network Center there are many Internet users in my country, and the Internet penetration rate of users is relatively high.
  • the popularity of Internet equipment has also had a huge impact on the catering industry. More and more people choose the Internet to order, comment and share food.
  • Text mining technology in the catering field is not widely used.
  • the data mining technology in the catering industry mostly focuses on using machine learning or deep learning models to model various click data of users, predict their hobbies and behaviors, and recommend more suitable stores or dishes.
  • a large amount of text data cannot be quickly processed manually, and the value contained in it has not yet been fully tapped.
  • most of them focus on judging the emotional tendency of the text, while ignoring more fine-grained analysis. For example, for the service of the catering business, the taste of the dishes, the environment of the store, and the comments for each dish.
  • the embodiment of the present application provides a method for analyzing catering data, the method comprising:
  • the embodiment of the present application provides a catering data analysis device, including:
  • a comment data analysis module configured to obtain comment data, perform analysis through a natural language analysis method, and obtain a first analysis result
  • a visualization chart generating module configured to determine a visualization configuration according to a visualization request, and obtain a visualization chart according to the visualization configuration
  • a visualization display module configured to visually display at least one of the first analysis results through the visualization configuration and the visualization chart.
  • the embodiment of the present application provides an electronic device, including a processor and a memory;
  • the memory is used to store programs
  • the processor executes the program to implement a catering data analysis method, wherein the catering data analysis method includes:
  • the embodiment of the present application provides a computer-readable storage medium, the storage medium stores a program, and the program is executed by a processor to implement a method for analyzing catering data, wherein the method for analyzing catering data includes :
  • the embodiment of the present application provides a catering data analysis method, device, electronic equipment and storage medium, by performing a variety of natural language analysis on the catering data generated by the catering merchants when selling catering products, to determine the customer's preference in the comment data , satisfaction and feedback; through visual configuration and visual display, it provides the required visual display method for catering businesses, and through comprehensive analysis and analysis of natural language, it can quickly understand consumers' concerns, needs, and the product itself. Insufficient points and problems related to store services; provide merchants with more accurate guidance on customer needs and satisfaction.
  • Fig. 1 is a schematic flowchart of the method of the embodiment of the present application.
  • FIG. 2 is a schematic flowchart of a method for analyzing catering data according to an embodiment of the present application.
  • Fig. 3 is a schematic flowchart of a semantic network analysis method according to an embodiment of the present application.
  • Fig. 4 is a schematic diagram of a semantic network analysis result of an embodiment of the present application.
  • Fig. 5 is a schematic flowchart of a sentiment analysis method according to an embodiment of the present application.
  • FIG. 6 is a schematic flow chart of a viewpoint extraction method according to an embodiment of the present application.
  • Fig. 7 is a schematic flowchart of the visualization configuration of the embodiment of the present application.
  • FIG. 8 is a schematic diagram of user portrait visualization according to an embodiment of the present application.
  • FIG. 9 is a schematic diagram of a visualized label extraction according to an embodiment of the present application.
  • Fig. 10 is a flow chart of a method for comparative analysis of catering data according to an embodiment of the present application.
  • FIG. 11 is a schematic diagram of a comparison and visualization of emotion index changes in an embodiment of the present application.
  • Fig. 12 is a schematic diagram of visual comparison of competing product labels in the embodiment of the present application.
  • Fig. 13 is a schematic diagram of an annual visualization report of a merchant's catering data according to an embodiment of the present application.
  • Fig. 14 is a diagram of a catering data analysis device according to an embodiment of the present application.
  • the application provides a catering data analysis method, device, electronic equipment, and storage medium.
  • the catering data analysis method includes: firstly obtaining comment data, performing analysis through a natural language analysis method, and obtaining the first analysis result; secondly, responding to the visualization request , determining a visualization configuration, obtaining a visualization chart according to the visualization configuration; finally, visually displaying at least one of the first analysis results through the visualization configuration and the visualization chart.
  • This application performs a variety of natural language analysis on the catering data generated by catering merchants when they sell catering products to determine customer preferences, satisfaction and feedback in the review data; through visual configuration and visual display, it provides all catering merchants
  • the visual display of demand can quickly understand consumers' concerns and needs, as well as the shortcomings of the product itself and the problems related to store services through comprehensive analysis and analysis of natural language; provide merchants with more accurate customer needs and satisfaction degree guidance.
  • AI artificial intelligence
  • the embodiments of the present application may acquire and process relevant data based on artificial intelligence technology.
  • artificial intelligence is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results. .
  • Artificial intelligence basic technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operation/interaction systems, and mechatronics.
  • Artificial intelligence software technology mainly includes computer vision technology, robotics technology, biometrics technology, speech processing technology, natural language processing technology, and machine learning/deep learning.
  • Natural language processing technology is an important direction in the field of computer science and artificial intelligence. It studies various theories and methods that can realize effective communication between humans and computers using natural language. Natural language processing is a science that combines linguistics, computer science, and mathematics. Research in this field will therefore involve natural language, the language that people use every day, so it is closely related to the study of linguistics, but has important differences. Natural language processing is not the general study of natural language, but the development of computer systems that can effectively realize natural language communication, especially the software systems. As such it is part of computer science.
  • Natural language processing is mainly used in machine translation, public opinion monitoring, automatic summarization, opinion extraction, text classification, question answering, text semantic comparison, speech recognition, Chinese OCR, etc.
  • the execution subject of the above-mentioned catering data analysis may be a computer device, and the computer device may be a terminal or a server.
  • the terminals mentioned here can be smartphones, tablet computers, laptop computers, desktop computers, vehicle-mounted computers, smart homes, wearable electronic devices, VR (Virtual Reality, virtual reality)/AR (Augmented Reality, augmented reality) devices etc.
  • the server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or it can provide cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud Communications, middleware services, domain name services, security services, content delivery network (Content Delivery Network, CDN), and cloud servers for basic cloud computing services such as big data and artificial intelligence platforms, etc.
  • the data in this embodiment of the application can be stored in a server, and the server can be an independent server, or it can provide cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, intermediate Cloud servers for basic cloud computing services such as mail service, domain name service, security service, content delivery network (Content Delivery Network, CDN), and big data and artificial intelligence platforms.
  • the server can be an independent server, or it can provide cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, intermediate Cloud servers for basic cloud computing services such as mail service, domain name service, security service, content delivery network (Content Delivery Network, CDN), and big data and artificial intelligence platforms.
  • CDN Content Delivery Network
  • FIG. 1 is a schematic structural diagram of an implementation environment provided by the present application.
  • the implementation environment includes a terminal 101 , a server 102 and a terminal 103 .
  • a visual software system is installed in the terminal 101, and the software system can perform visual display and configuration.
  • the editing interface of the software system is a visual user interface, in which the operation area is divided, and the divided areas include but not limited to the visual display area and the visual editing area.
  • the terminal 103 is equipped with a software system for comments, which can edit and send comments.
  • terminal 101 and terminal 103 can be any electronic product that can perform human-computer interaction in one or more ways such as keyboard, touch pad, touch screen, remote control, voice interaction or handwriting equipment.
  • the product can receive the user's target operation instructions through its visual operation interface, and perform real-time visual display of the results of the operation instructions.
  • the terminal 101 and the terminal 103 can be a personal computer (Personal Computer, PC), a mobile phone, a smart phone, a personal digital assistant (Personal Digital Assistant, PDA), a wearable device, a handheld computer (PPC) (Pocket PC), tablet, etc.
  • the terminal 101 and the terminal 103 can establish a communication connection with the server 102 through a network protocol, and comment data is stored in the database of the server 102, and the comment data passes through the customer terminal.
  • the server 102 may push corresponding visualization results to the terminal 101 according to different visualization instructions sent by the terminal 101 .
  • the comment data collection code is stored in the server 102 for continuous collection of comment data.
  • the server can be an independent server, or a server cluster composed of several servers, or a cloud computing service center; It has a data storage function, and the realization of the storage function can use a local database or a cloud database.
  • the complete process of catering data analysis and processing is as follows: on the editing interface of the terminal 103, the comment data edited by the user is obtained and sent to the server 102 , the server 102 acquires the comment data to be analyzed, and executes the analysis using a natural language analysis method to obtain the corresponding analysis results; on the editing interface of the terminal 101, that is, the visual user operation interface, firstly obtain the visualization request, determine the visualization configuration, and according to the visualization configuration Generate a visual chart, obtain the analysis result of the natural language analysis from the server 102, and display the analysis result on the editing interface through the visual chart.
  • this embodiment of the present application is only illustratively illustrating a possible implementation environment and a possible implementation result through FIG. 1 , that is, the user sends a visualization request through the terminal 101, and the server 102 , package the corresponding comment data and feed it back to the terminal 101 for visual editing or visual display.
  • the comment data analysis method can also be implemented only by the terminal 101 equipped with a comment data processing software system.
  • the user can follow The data cached in the terminal is operated offline, and the visual analysis results are obtained according to the comment data visualization program built in the terminal 101; after reconnecting to the server 102, the terminal 101 is cached locally in the visual analysis results and related continuity relationships Upload and update to the server 102 for data storage and update; as another example, in another embodiment, the server 102 obtains the comment data sent by the terminal 101, the server 102 sends the comment data according to the visualization request sent by the terminal 101, and the terminal 101 receives After commenting on the data, the analysis is performed by executing a natural language analysis method to obtain a visual analysis result, and through the visual configuration, a visual chart is generated and displayed on the editing interface of the terminal 101 .
  • the embodiment of the present application does not limit specific application scenarios, and the above-mentioned application scenario in FIG. 1 is only used as an example for illustration.
  • the embodiment of the present application provides a process flow of a catering data analysis method, which can be implemented by applying the terminal 101 in Figure 1; or, this method can also be implemented in any other data processing capable device or equipment.
  • the method specifically includes but not limited to steps S100-S300.
  • step S100 comment data is obtained, and analysis is performed by a natural language analysis method to obtain a first analysis result.
  • the comment data in the embodiment is the customer's comment on consuming the catering product, and is the customer's emotional feedback on the catering product during consumption.
  • the comment data generated by customers is acquired and stored when it is generated, or it can be called or crawled through an interface from a third-party platform.
  • the analysis performed by the natural language analysis method in the embodiment is to use different natural language methods to analyze the comment data correspondingly, and the natural language method is to analyze human language, obtain the emotions included in the comment data, and analyze these emotions Analyze and classify.
  • the comment data is the evaluation text of human subjective behavior, therefore, opinion mining needs to be performed on the comment data through natural language methods.
  • the method for analyzing emotion in natural language includes a semantic network analysis method, a sentiment analysis method, and an opinion extraction method.
  • a schematic flowchart of a semantic network analysis method is provided.
  • the process of the semantic network analysis method may include but not limited to steps S310-S330:
  • step S320 perform nonsense word filtering and feature extraction to high-frequency words, obtain the feature vector of high-frequency words;
  • step S310 include the following comment fields "I think the service of businessman A is good” and respectively in two comment data "I think merchant B's service needs to be improved", some features of the field “I think merchant A's service is good” are obtained through meaningless word filtering and feature extraction are "I", “Think”, “Service”, “Good”, The characteristics of the field “I think the service of merchant B needs to be improved” are "I”, “I feel”, “service” and “need to be improved”.
  • step S330 Create a co-occurrence matrix, input the feature vectors into the co-occurrence matrix, and obtain a semantic network analysis result.
  • the fields "I think the service of merchant A is good” and “I think the service of merchant B needs to be improved” are respectively input into the rows and columns of the sharing matrix, and the number of the same characteristics is counted.
  • FIG. 4 when a number of different types of merchants such as Sichuan and Hunan cuisine, cake dessert, Southeast Asian cuisine, hot pot, Beijing cuisine and Shandong cuisine, dinner banquet, Japanese and Korean cuisine, barbecue barbecue, soup pot, dessert drink, Western food, fragrant Grilled fish in a pot, snacks, fast food, buffet, etc., when the above steps S310-S330 are performed, a schematic diagram of the semantic network analysis results shown in Figure 4 is obtained.
  • the left side of Figure 4 represents the words analyzed by the network, and the circles The size indicates the degree of association between different semantic words and other semantic words.
  • the circle of "service” is the largest, indicating that "service” is the most associated high-frequency word.
  • a schematic flowchart of a sentiment analysis method is provided.
  • the process of the semantic network analysis method may include but not limited to steps S510-S330:
  • S510 perform sentiment word analysis on the comment data through the sentiment dictionary, and obtain negative words, positive words, degree adverbs and degree symbols; for each word, it is necessary to quantify the intensity of emotion included in it through the sentiment dictionary, and first judge its Whether it is an emotional word, for example, for the comment data including the emotional words “this dish is very delicious” and “the service of this store is relatively poor", through emotional dictionaries such as positive dictionaries, negative dictionaries and degree dictionaries, positive The word “good”, the negative word “poor”, the adverbs of degree "very” and “comparative".
  • step S520 assigning values to negative words and positive words respectively, and assigning weights and weight calculation methods for degree adverbs and degree symbols; for example, based on the embodiment of step S510, the positive word "good” is assigned a value of 1, and the negative word “poor” is -1, the weights of the degree adverbs "very” and “comparative” are set to 0.75 and -0.25 respectively, and the weight calculation method is the product.
  • the weight calculation method is the product.
  • a schematic flow chart of a viewpoint extraction method is provided.
  • the procedure of the viewpoint extraction method may include but not limited to steps S610-S630:
  • S610 perform word segmentation on the comment data, create a graph, and add the word obtained by word segmentation as a node to the graph; for example, the word segmentation divides the text into several words, performs word segmentation and part-of-speech tagging during segmentation, and filters out disabled Words, only keep the words that specify the part of speech, such as nouns, verbs, and adjectives, and the obtained words are used as nodes, such as "the taste of this dish is great", and “dishes", “taste”, and “sticks” are obtained through examples and other nodes, and build a graph, where the graph is used to show the resulting words and their relationships.
  • the word segmentation divides the text into several words, performs word segmentation and part-of-speech tagging during segmentation, and filters out disabled Words, only keep the words that specify the part of speech, such as nouns, verbs, and adjectives, and the obtained words are used as nodes, such as "the taste of this dish is great", and
  • S620 identify the relationship between each word in the graph through the window, and draw a connection between the words that have the relationship;
  • the size of the window is intercepted, and each word is used as a node of the candidate keyword graph, and the words in each segment of the intercepted text are used as adjacent edges, and a connection line is drawn between the related words to construct the above candidate keyword graph.
  • step S630 sort the words according to the number of connections of each node, and obtain an opinion extraction result of the comment data. For example, iterate the candidate keyword map obtained in step S620, initialize the weight of each node, and after the set number of iterations reaches stability, sort the node weights in reverse order to obtain an opinion extraction method, in which the opinion is the key word.
  • the visualization request is an instruction to obtain the first analysis result and perform visualization.
  • the instruction includes the first analysis result to be visualized and its corresponding visualization configuration.
  • a merchant needs to visualize multiple data, so first obtain the visualization configuration , to determine the data requirements of the first analysis result required, execute the data pull of the first analysis result, and at the same time, obtain the form to be displayed in the visualization according to the visualization configuration, such as radar charts, emotional change charts, etc., and create corresponding visualization charts.
  • FIG. 7 a schematic flow diagram of a visualization configuration is provided.
  • the flow of the visualization configuration may include but not limited to steps S710-S730:
  • Visual format For example, a merchant wants to view user portraits in November, and the display method is a radar chart, so the visualization format is the data representation to be visualized, the visualization chart type is a radar chart, and the visualization time is November 1-11 On March 31, the types of visualization results are sentiment analysis results and semantic network analysis results, where the data representation can be in the form of percentage values or column charts.
  • calling the visualization chart interface to generate a visualization chart according to the visualization format and the visualization chart may use calling a visualization control or calling a visualization interface.
  • the embodiment provides a schematic diagram of user portrait visualization.
  • the merchant wants to view the user portrait in November, and the display method is a radar chart. Therefore, the visualization format is the data representation to be visualized, and the visualization chart type is a radar chart.
  • the visualization time is the comment data generated from November 1st to November 31st, and the types of visualization results are sentiment analysis results and semantic network analysis results.
  • FIG 8 which provides a visual schematic diagram of user portraits.
  • taste, Star rating, service, variety, and environment are the results of semantic network analysis, and the corresponding ratios include taste 3.95, star rating 3.94, service 3.95, variety 4.04, and environment 3.93.
  • These corresponding sentiment analysis results in one embodiment, are visualized
  • the data representation form is 5*positive emotion/(positive emotion+negative emotion).
  • the first analysis result may be expressed as an analysis result of a natural language analysis method of a merchant's comment data; wherein, in a feasible embodiment, the visual configuration and the visual chart to visually display at least one first analysis result include According to the visualization instructions, at least one analysis result of the merchant's comment data is displayed visually, and displayed according to the visualization configuration and analysis results, such as label extraction, opinion clustering, word cloud display, emotional change, and emotional polarity.
  • the embodiment provides a visual schematic diagram of label extraction, wherein the label extraction includes extracting based on the above-mentioned sentiment analysis results and viewpoints, and this embodiment provides a visual schematic diagram of single label extraction and business comment mining.
  • the figure follows the direction of the arrow to complete the visualization process of the label.
  • the left side of the figure includes the customer's comment and the comment opinion mining selected by the merchant.
  • This embodiment obtains a single comment label by performing natural language processing such as a network analysis method, a sentiment analysis method or a viewpoint extraction method on a single comment. After all the tags of the merchant are counted, the tag extraction results shown on the lower right are obtained, so as to fully mine the sales of the merchant from multiple angles.
  • FIG. 10 illustrates a flow chart of a method for comparative analysis of catering data, which specifically includes but not limited to steps S1010-S1020.
  • S1010 Obtain a second analysis result of at least one other merchant according to the merchant's catering type and preset rules, and visually compare and display the first analysis result and the second analysis result through a visual chart.
  • other merchants that compete with the merchant are determined through the merchant’s catering type and preset rules. For example, if the merchant’s catering type is the merchant’s main dish is Hunan cuisine, based on this catering type to find other merchants whose food is Hunan cuisine, the preset rule can be at least one of geographical distance, order quantity or user visits.
  • this embodiment uses geographical location, order quantity and user visits For example, find Hunan cuisine restaurants in a certain district of a certain city, and the number of orders is greater than 1,000, and the number of user visits is greater than 10,000.
  • the preset rule is a flexible configuration, further, it can also be based on, for example, the ratio of daily visits/transactions, or the number of new users.
  • the second analysis result is that other merchants based on the first
  • the visualization configuration and visualization chart of the analysis results are obtained. Further, the analysis results of all merchants are visualized using the same configuration table to provide merchants with visual customer demand and satisfaction guidance.
  • the suggested text compares the sentiments of the same words and keywords obtained by at least one of the semantic network analysis method, sentiment analysis method or opinion extraction method between the first analysis result and the second analysis result to determine the emotional difference, according to the emotional difference Automatically generate suggestions or evaluation texts.
  • the semantic network analysis method obtains the word "service”, and the positive comments "service” are respectively obtained by performing sentiment analysis on the comments of the merchants including “service” ( 617)” and “Service (98)", through the comparison, it is determined that the service of the merchant who positively commented on “Service (617)" is better than the merchant of "Service (98)", and for the merchant with poor service, generate the following suggested text " Consumers pay more attention to services, so improving the service level has a certain effect on increasing praise", in which the suggested text is generated by adding fixed-format text and variable text, such as the above-mentioned "service” is a variable text, and “consumers to __ Pay more attention, so increasing the degree of __ has a certain effect on increasing praise" is a fixed-format text.
  • the method of generating the suggested text is not limited to the above-mentioned implementations, and methods such as machine learning can also be used. Different suggestions are used as the input of the machine learning model, and the output
  • FIG. 11 a schematic diagram of a comparative visualization of sentiment index changes is provided, which illustrates the changes of sentiment indices of two merchants C1 and C2 in different time periods.
  • the abscissa in the figure is time, and the ordinate is is the emotional value.
  • merchants C1 and C2 in the figure have the same catering type.
  • the merchant C1 with the highest sentiment index and the merchant C2 with the lowest sentiment index in the time period are displayed, that is, merchant C1 is the waveform of the emotional index change at the top, that is, the merchant C2 is the waveform of the emotional index change at the bottom of C1.
  • the emotional visualization comparison is performed on the comment data of customers of different merchants.
  • FIG. 12 a schematic diagram of a visual comparison of competing product labels is provided.
  • different labels of three merchants D1, D2 and D3 and their sentiment statistics are illustrated.
  • semantic The network analysis method determines the semantic analysis results of merchants D1, D2 and D3.
  • the semantic analysis results of merchant D1 are taste, service, environment, delicious, fresh, attitude, taste, price, queue, etc., expensive, poor, Cheap, hygienic, cost-effective, high quality and low price, and then perform sentiment analysis method on the comment data corresponding to the semantic words, that is, taste 896, service 486, environment 386, delicious 307, freshness 294, attitude 167, taste 117, price 94, Visual analysis results such as queuing 83, waiting 73, expensive 71, poor 62 cheap 62, hygienic 46, cost-effective 46, high quality and low price 13, etc.
  • this embodiment also shows that according to the user's selection of positive comments and negative comments, for example, the merchant in D2 has negative comments, and D1 and D3 include comprehensive visualization results of positive comments and negative comments.
  • FIG 13 illustrates a schematic diagram of an annual visualization report on a merchant's catering data, which includes five pages P1, P2, P3, P4, and P5,
  • P1 is the home page of the 2020 annual visualization report on the merchant's catering data
  • P2 is a visual diagram showing the analysis results of the word cloud
  • P3 is the best comment of a user (that is, label extraction)
  • P4 is the sorting result of popular dishes
  • P5 is the generated suggestion or evaluation text.
  • P1 is a visualization request from a merchant, and the request is an annual report. It can be known that the visualization format, visualization chart type, visualization time, and visualization result type are respectively obtained.
  • the visualization format is multi-page Display
  • the type of visualization chart is "annual report”
  • the visualization time is from 2020.1.1 to 2020.12.31
  • the types of visualization results include comprehensive analysis results as shown in P1 to P5.
  • the analysis results are analyzed through semantic network analysis methods, sentiment analysis methods or Multiple generation in the opinion extraction method, such as the word cloud display in P2, which obtains semantic words through the semantic network analysis method, and performs opinion extraction with the opinion extraction method, obtains the word cloud analysis result, and visualizes it on the P2 page;
  • P5 by performing sentiment analysis on the comments of multiple merchants, "service” and “dish” are obtained respectively. After comparing with other merchants, the sentiment analysis results of these two words are poor, and "service” and “dish” are generated. suggested or evaluated text.
  • the embodiment of the present application also provides a catering data analysis and analysis device, which includes a comment data analysis module 1401, a visual chart generation module 1402 and a visual display module 1403;
  • the comment data analysis module is used to obtain comment data, perform analysis through natural language analysis methods, and obtain the first analysis result
  • the visualization chart generation module is used to determine the visualization configuration according to the visualization request, and obtain the visualization chart according to the visualization configuration
  • a display module configured to visually display at least one first analysis result through a visual configuration and a visual chart.
  • the device in this embodiment can implement any of the aforementioned catering data analysis methods, that is, obtain review data, and use natural language analysis methods to Executing the analysis; determining the visualization configuration, obtaining a visualization chart according to the visualization configuration; visually displaying the analysis results through the visualization configuration and the visualization chart.
  • This application performs a variety of natural language analysis on the catering data generated by catering merchants when they sell catering products to determine customer preferences, satisfaction and feedback in the review data; through visual configuration and visual display, it provides all catering merchants
  • the visual display of demand can quickly understand consumers' concerns and needs, as well as the shortcomings of the product itself and the problems related to store services through comprehensive analysis and analysis of natural language; provide merchants with more accurate customer needs and satisfaction degree guidance.
  • the embodiment of the present application also provides an electronic device, the electronic device includes a processor and a memory;
  • the memory stores a program
  • the processor executes the program to implement a method for analyzing catering data, wherein the method for analyzing catering data includes: acquiring comment data, performing analysis through a natural language analysis method, and obtaining a first analysis result; in response to a visualization request, determining a visualization configuration Obtaining a visualization chart according to the visualization configuration; visually displaying at least one of the first analysis results through the visualization configuration and the visualization chart.
  • this electronic device has the function of carrying and running the software system of the catering data analysis provided by the embodiment of the present application, for example, personal computer (Personal Computer, PC), mobile phone, smart phone, personal digital assistant (Personal Digital Assistant, PDA) , Wearable devices, Pocket PC (Pocket PC), Tablet PC, etc.
  • the embodiment of the present application also provides a computer-readable storage medium, the storage medium stores a program, and the program is executed by a processor to implement a method for analyzing catering data, wherein the method for analyzing catering data includes: acquiring comment data, Perform analysis by a natural language analysis method to obtain a first analysis result; in response to a visualization request, determine a visualization configuration, and obtain a visualization chart according to the visualization configuration; use the visualization configuration and the visualization chart for at least one of the first Analysis results are visualized.
  • the computer-readable storage medium may be non-volatile or volatile.
  • the functions/operations noted in the block diagrams may occur out of the order noted in the operational diagrams.
  • two blocks shown in succession may, in fact, be executed substantially concurrently or the blocks may sometimes be executed in the reverse order, depending upon the functionality/operations involved.
  • the embodiments presented and described in the flowcharts of this application are provided by way of example for the purpose of providing a more comprehensive understanding of the technology. The disclosed methods are not limited to the operations and logical flow presented herein. Alternative embodiments are contemplated in which the order of various operations is changed and in which sub-operations described as part of larger operations are performed independently.
  • the embodiment of the present application also discloses a computer program product or computer program, where the computer program product or computer program includes computer instructions, and the computer instructions are stored in a computer-readable storage medium.
  • the processor of the computer device can read the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes a method for analyzing catering data, wherein the method for analyzing catering data includes: obtaining comment data , performing an analysis by a natural language analysis method to obtain a first analysis result; in response to a visualization request, determining a visualization configuration, and obtaining a visualization chart according to the visualization configuration; performing at least one of the first analysis by using the visualization configuration and the visualization chart The results are displayed visually.
  • the functions described above are realized in the form of software function units and sold or used as independent products, they can be stored in a computer-readable storage medium.
  • the technical solution of the present application is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including Several instructions are used to make a computer device (which may be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the methods described in the various embodiments of the present application.
  • the aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disk or optical disc, etc., which can store program codes. .
  • a "computer-readable medium” may be any device that can contain, store, communicate, propagate or transmit a program for use in or in conjunction with an instruction execution system, device or device.
  • computer-readable media include the following: electrical connection with one or more wires (electronic device), portable computer disk case (magnetic device), random access memory (RAM), Read Only Memory (ROM), Erasable and Editable Read Only Memory (EPROM or Flash Memory), Fiber Optic Devices, and Portable Compact Disc Read Only Memory (CDROM).
  • the computer-readable medium may even be paper or other suitable medium on which the program can be printed, as it may be possible, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or other suitable processing if necessary.
  • the program is processed electronically and stored in computer memory.
  • each part of the present application may be realized by hardware, software, firmware or a combination thereof.
  • various steps or methods may be implemented by software or firmware stored in memory and executed by a suitable instruction execution system.
  • a suitable instruction execution system For example, if implemented in hardware, as in another embodiment, it can be implemented by any one or combination of the following techniques known in the art: Discrete logic circuits, ASICs with suitable combinational logic gates, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), etc.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Computational Linguistics (AREA)
  • General Health & Medical Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

本申请提供一种餐饮数据分析方法、装置、电子设备及存储介质，该餐饮数据分析方法包括：获取评论数据，通过自然语言分析方法执行分析；确定可视化配置，根据可视化配置得到可视化图表；通过可视化配置及可视化图表对分析结果进行可视化显示。本申请通过对餐饮商家在售卖餐饮产品时所产生的餐饮数据执行多种自然语言分析，确定评论数据中顾客的偏好、满意度和意见反馈；通过可视化配置及可视化显示，为餐饮商家提供了所需求的可视化显示方式，通过自然语言的综合分析分析可以快速地了解消费者的关注点、需求点，以及商品本身的不足点和店铺服务相关的问题点；为商家提供更精准的顾客需求和满意度指导。

Description

餐饮数据分析方法、装置、电子设备及存储介质
本申请要求于2022年2月22日提交中国专利局、申请号为202210163511.X，发明名称为“餐饮数据分析方法、装置、电子设备及存储介质”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及计算机、人工智能及自然语言处理技术领域，尤其涉及一种餐饮数据分析方法、装置、电子设备及存储介质。
背景技术
目前，线上市场也成为企业新的快速增长点。据互联网络中心的统计显示，我国网络用户众多，用户互联网普及率较高，互联网设备的普及对餐饮行业也产生了巨大的影响，相较之前的店内堂食，口口相传进行店铺宣传，越来越多的人选择了网络这种方式来点餐、评论和分享美食。
技术问题
以下是发明人意识到的现有技术的技术问题：
(1)餐饮领域的文本挖掘技术应用不广泛。当前，餐饮行业的数据挖掘技术多集中于利用机器学习或深度学习的模型对用户的各类点击数据进行建模，预测其爱好和行为，推荐更合适的店铺或菜品。而随着用户在线评论习惯的养成，大量文本数据无法快速人工处理，其中蕴含的价值还未被完全挖掘出来。需要利用计算机技术从海量数据中自动总结观点，为餐饮商家、餐饮平台为消费者持续优化产品提供依据。当前对于文本评论数据的分析中，多集中于对文本情感倾向进行判断，而忽视了更加细粒度的分析。比如针对餐饮商家的服务、菜品的口味、店铺的环境，针对每道菜的点评。
(2)针对商家方的服务形式单一。餐饮平台对餐饮商家提供的数据分析大多仅有每日各类数据如访问量、下单量、成交量、好评/差评条数等的简单趋势展示，而缺少对文本数据深度挖掘后的分析，也不会提供具体建议。这忽视了文本数据中的价值，餐饮商家也无法从其中获得更为专业的改进方法，对商家的销售和消费者的服务感受没有提升。
技术解决方案
第一方面，本申请实施例提供了一种餐饮数据分析方法，该方法包括：
获取评论数据，通过自然语言分析方法执行分析，得到第一分析结果；
响应于可视化请求，确定可视化配置，根据所述可视化配置得到可视化图表；
通过所述可视化配置及所述可视化图表对至少一种所述第一分析结果进行可视化显示。
第二方面，本申请实施例提供了一种餐饮数据分析装置，包括：
评论数据分析模块，用于获取评论数据，通过自然语言分析方法执行分析，得到第一分析结果；
可视化图表生成模块，用于根据可视化请求，确定可视化配置，根据所述可视化配置得到可视化图表；
可视化显示模块，用于通过所述可视化配置及所述可视化图表对至少一种所述第一分析结果进行可视化显示。
第三方面，本申请实施例提供了一种电子设备，包括处理器以及存储器；
所述存储器用于存储程序；
所述处理器执行所述程序实现一种餐饮数据分析方法，其中，所述餐饮数据分析方法包括：
获取评论数据，通过自然语言分析方法执行分析，得到第一分析结果；
响应于可视化请求，确定可视化配置，根据所述可视化配置得到可视化图表；
通过所述可视化配置及所述可视化图表对至少一种所述第一分析结果进行可视化显示。第四方面，本申请实施例提供了一种计算机可读存储介质，所述存储介质存储有程序，所述程序被处理器执行实现一种餐饮数据分析方法，其中，所述餐饮数据分析方法包括：
获取评论数据，通过自然语言分析方法执行分析，得到第一分析结果；
响应于可视化请求，确定可视化配置，根据所述可视化配置得到可视化图表；
通过所述可视化配置及所述可视化图表对至少一种所述第一分析结果进行可视化显示。
有益效果
本申请的实施例提供了一种餐饮数据分析方法、装置、电子设备及存储介质，通过对餐饮商家在售卖餐饮产品时所产生的餐饮数据执行多种自然语言分析，确定评论数据中顾客的偏好、满意度和意见反馈；通过可视化配置及可视化显示，为餐饮商家提供了所需求的可视化显示方式，通过自然语言的综合分析分析可以快速地了解消费者的关注点、需求点，以及商品本身的不足点和店铺服务相关的问题点；为商家提供更精准的顾客需求和满意度指导。
附图说明
本申请的上述和/或附加的方面和优点从结合下面附图对实施例的描述中将变得明显和容易理解，其中：
图1是本申请实施例的方法的流程示意图。
图2是本申请实施例的餐饮数据分析方法的流程示意图。
图3是本申请实施例的语义网络分析方法的流程示意图。
图4是本申请实施例的语义网络分析结果示意图。
图5是本申请实施例的情感分析方法的流程示意图。
图6是本申请实施例的观点抽取方法的流程示意图。
图7是本申请实施例的可视化配置的流程示意图。
图8是本申请实施例的用户画像可视化示意图。
图9是本申请实施例的标签抽取可视化示意图。
图10是本申请实施例的餐饮数据对比分析方法的流程图。
图11是本申请实施例的情感指数变化对比可视化的示意图。
图12是本申请实施例的竞品标签的可视化对比示意图。
图13是本申请实施例的商户餐饮数据年度可视化报告示意图。
图14是本申请实施例的餐饮数据分析分析装置图。
本发明的实施方式
下面详细描述本申请的实施例，所述实施例的示例在附图中示出，其中自始至终相同或类似的标号表示相同或类似的元件或具有相同或类似功能的元件。在后续的描述中，使用用于表示元件的诸如“模块”、“部件”或“单元”的后缀仅为了有利于本申请的说明，其本身没有特有的意义。因此，“模块”、“部件”或“单元”可以混合地使用。“第一”、“第二”等只是用于区分技术特征为目的，而不能理解为指示或暗示相对重要性或者隐含指明所指示的技术特征的数量或者隐含指明所指示的技术特征的先后关系。在本后续的描述中，对方法步骤的连续标号是为了方便审查和理解，结合本申请的整体技术方案以及各个步骤之间的逻辑关系，调整步骤之间的实施顺序并不会影响本申请技术方案所达到的技术效果。下面通过参考附图描述 的实施例是示例性的，仅用于解释本申请，而不能理解为对本申请的限制。
本申请提供了一种餐饮数据分析方法、装置、电子设备及存储介质，餐饮数据分析方法包括：首先获取评论数据，通过自然语言分析方法执行分析，得到第一分析结果；其次，响应于可视化请求，确定可视化配置，根据所述可视化配置得到可视化图表；最后，通过所述可视化配置及所述可视化图表对至少一种所述第一分析结果进行可视化显示。本申请通过对餐饮商家在售卖餐饮产品时所产生的餐饮数据执行多种自然语言分析，确定评论数据中顾客的偏好、满意度和意见反馈；通过可视化配置及可视化显示，为餐饮商家提供了所需求的可视化显示方式，通过自然语言的综合分析分析可以快速地了解消费者的关注点、需求点，以及商品本身的不足点和店铺服务相关的问题点；为商家提供更精准的顾客需求和满意度指导。
本申请实施例可以基于人工智能技术对相关的数据进行获取和处理。其中，人工智能(Artificial Intelligence，AI)是利用数字计算机或者数字计算机控制的机器模拟、延伸和扩展人的智能，感知环境、获取知识并使用知识获得最佳结果的理论、方法、技术及应用系统。
人工智能基础技术一般包括如传感器、专用人工智能芯片、云计算、分布式存储、大数据处理技术、操作/交互系统、机电一体化等技术。人工智能软件技术主要包括计算机视觉技术、机器人技术、生物识别技术、语音处理技术、自然语言处理技术以及机器学习/深度学习等几大方向。
自然语言处理技术是计算机科学领域与人工智能领域中的一个重要方向。它研究能实现人与计算机之间用自然语言进行有效通信的各种理论和方法。自然语言处理是一门融语言学、计算机科学、数学于一体的科学。因此，这一领域的研究将涉及自然语言，即人们日常使用的语言，所以它与语言学的研究有着密切的联系，但又有重要的区别。自然语言处理并不是一般地研究自然语言，而在于研制能有效地实现自然语言通信的计算机系统，特别是其中的软件系统。因而它是计算机科学的一部分。
自然语言处理主要应用于机器翻译、舆情监测、自动摘要、观点提取、文本分类、问题回答、文本语义对比、语音识别、中文OCR等方面。
上述所提及的餐饮数据分析的执行主体可以是计算机设备，该计算机设备可以是终端或者服务器。此处所提及的终端可以是智能手机、平板电脑、笔记本电脑、台式电脑、车载计算机、智能家居、可穿戴电子设备、VR(Virtual Reality，虚拟现实)/AR(Augmented Reality，增强现实)设备等等；服务器可以是独立的物理服务器，也可以是多个物理服务器构成的服务器集群或者分布式系统，还可以是提供云服务、云数据库、云计算、云函数、云存储、网络服务、云通信、中间件服务、域名服务、安全服务、内容分发网络(Content Delivery Network，CDN)、以及大数据和人工智能平台等基础云计算服务的云服务器，等等。
需要说明的是，本申请实施例的数据可以保存在服务器中，服务器可以是独立的服务器，也可以是提供云服务、云数据库、云计算、云函数、云存储、网络服务、云通信、中间件服务、域名服务、安全服务、内容分发网络(Content Delivery Network，CDN)、以及大数据和人工智能平台等基础云计算服务的云服务器。
图1是本申请提供的一种实施环境的结构示意图。如图1所示，在该实施环境中，包括了终端101、服务器102及终端103。在终端101中搭载了可视化的软件系统，该软件系统可以进行可视化的显示和配置。软件系统的编辑界面为可视化用户界面，在该用户界面中进行操作区域的划分，划分得到区域包括但不限于可视化显示区域和可视化编辑区域。终端103搭载了用于评论的软件系统，该软件系统可以进行评论的编辑及发送。在该实施环境中，终端101和终端103可以是任何一种可通过键盘、触摸板、触摸屏、遥控器、语音交互或手写设备等一种或多种方式进行人机交互的电子产品，该电子产品可以通过其可视化操作界面接收用户的目标操作指令，并对操作指令的结果进行实时的可视化显示。示例性地，如图1所示，该终端101和终端103可以是个人计算机(Personal Computer，PC)、手机、智能手机、个人数字助手(Personal Digital Assistant，PDA)、可穿戴设备、掌上电脑PPC(Pocket PC)、平板 电脑等。
此外，在图1所示的实施环境中，终端101和终端103可以通过网络协议与服务器102建立通信连接，在服务器102的数据库中存储有评论数据，该评论数据通过顾客终端。服务器102可以根据终端101所发送的不同的可视化指令，推送相应的可视化结果至终端101。在服务器102中存储有评论数据采集代码，用于进行连续性地评论数据的采集。在该实施环境中，服务器可以采用独立的服务器，或者是由若干台服务器组成的服务器集群，又或者是云计算服务中心；服务器的形式不限于上述举例描述的形式，可以理解的是，服务器还具有数据存储功能，而存储功能的实现可以采用本地数据库或者是云端的数据库。
以某一具体的餐饮数据分析过程为例，在图1所示的实施环境中，完整的餐饮数据分析处理的过程为：在终端103的编辑界面，获取用户编辑的评论数据并发送至服务器102，服务器102获取待分析的评论数据，并执行自然语言分析方法执行分析，得到相应的分析结果；在终端101的编辑界面，即可视化用户操作界面，首先获取可视化请求，确定可视化配置，根据可视化配置生成可视化图表，从服务器102获取已进行自然语言分析的分析结果，通过可视化图表将分析结果在该编辑界面进行显示。
应当理解的是，本申请实施例仅是通过图1示例性地说明可以实施的一种实施环境以及一种可能性的实施结果，即用户通过终端101发送了可视化请求，服务器102根据该可视化请求，将对应的评论数据进行打包并反馈至终端101进行可视化编辑或者可视化显示。而在本申请实施例另一些应用场景中，评论数据分析方法也可以仅由搭载了评论数据处理软件系统的终端101实现，例如：在终端101断开与服务器102的通信连接之后，用户可以根据终端中缓存的数据进行离线操作，根据内置在终端101中的评论数据可视化程序，得到可视化分析结果；待重新连接至服务器102后，将终端101缓存在本地的可视化分析结果以及相关的连续性关系上传更新至服务器102进行数据的存储与更新；又如，在另一实施例中，服务器102获取终端101发送的评论数据，服务器102根据终端101发送的可视化请求，发送评论数据，终端101在接收评论数据后通过执行自然语言分析方法执行分析，得到可视化分析结果，通过可视化配置，生成可视化图表并显示于终端101的编辑界面。本申请实施例不对具体应用场景进行限定，上述的图1的应用场景仅仅作为示例性地说明。
如图2所示，本申请实施例提供了一种餐饮数据分析方法的流程，该流程方法可以应用于图1中的终端101来实现；或者，该方法也可以在其他任意具有数据处理能力的装置或设备上执行。参考图2，该方法具体包括但不限于步骤S100-S300。
步骤S100，获取评论数据，通过自然语言分析方法执行分析，得到第一分析结果。
其中，实施例中的评论数据是顾客在消费餐饮产品的评论，是消费时顾客对餐饮产品的情感反馈。评论数据的获取可以有多种方式，首先，顾客产生的评论数据在产生时即进行获取和存储，或者，可以通过从第三方平台进行接口调用或者爬取。
其中，实施例中通过自然语言分析方法执行分析是采用不同的自然语言方法对评论数据进行对应的分析，而自然语言方法是对人类语言进行分析，得到评论数据中包括的情绪，并对这些情绪进行解析和分类。实施例中，评论数据是人主观行为的评估文本，因此，需要通过自然语言方法对评论数据执行观点挖掘。实施例中，对情感进行自然语言分析的方法如语义网络分析方法、情感分析方法及观点抽取方法等。
示例性地，如图3所示，提供了一种语义网络分析方法的流程示意图，示例性地，在实施例中，语义网络分析方法的过程可以包括但不限于步骤S310-S330：
S310，提取评论数据中使用频率或出现次数超过预设值的高频词；其中，高频词的使用频率或出现次数是对评论数据的文本中的相同词进行搜索，直至完成所有文本的搜索，将相同或相似的总出现次数进行统计，基于预设值判断所搜索的次是否为高频词，例如，一条评论中，“我认为商家A的服务好”出现了10次，超过预设值5次的要求，即将“我认为商家A的服务好”作为高频词；
S320，对高频词执行无意义词过滤及特征提取，得到高频词的特征向量；基于步骤S310，例如，在两条评论数据中分别包括如下评论字段“我认为商家A的服务好”和“我觉得商家B的服务有待改进”，通过无意义词过滤及特征提取得到字段“我认为商家A的服务好”的一些特征为“我”、“认为”、“服务”、“好”，而“我觉得商家B的服务有待改进”字段的特征为“我”、“觉得”、“服务”、“有待改进”。
S330，创建共现矩阵，将特征向量输入至共现矩阵，得到语义网络分析结果。以步骤S320得到的特征，将字段“我认为商家A的服务好”和“我觉得商家B的服务有待改进”分别输入至共享矩阵的行和列，统计相同特征的数量，如两个字段均存在“服务”，将“服务”这个词作为语义网络分析的结果。
示例性地，参考图4，在对多个不同类型的商家如川湘菜、蛋糕甜品、东南亚菜、火锅、京菜鲁菜、聚餐宴请、日韩料理、烧烤烤肉、汤锅、甜点饮品、西餐、香锅烤鱼、小吃快餐、自助餐等多个执行如上述步骤S310～S330的处理时，得到如图4说展示的语义网络分析结果示意图，图4中左侧表示网络分析的词，而图中圆圈大小表示不同语义词与其他语义词的关联程度，如图4中“服务”的圆圈最大，表示“服务”为存在关联最多的高频词。
示例性地，如图5所示，提供了一种情感分析方法的流程示意图，示例性地，在实施例中，语义网络分析方法的过程可以包括但不限于步骤S510-S330：
S510，通过情感词典对评论数据执行情感词分析，得到消极词、积极词、程度副词及程度符号；对于每个词，需要通过情感词典对其所包括的情感强烈程度进行量化，首先要判断其是否为情感词，例如，对于包括有情感的词“这道菜非常好吃”和“这家店的服务比较差”的评论数据，通过情感词典如积极词典、消极词典及程度词典，得到积极词“好”，消极词“差”，程度副词“非常”及“比较”。
S520，对消极词和积极词分别赋值，并为程度副词及程度符号分配权重和权重计算方法；例如，基于步骤S510的实施例，将积极词“好”赋值为1，消极词“差”为-1，程度副词“非常”及“比较”的权重分别设置为0.75和-0.25，权重计算方法为乘积，通过对情感词进行赋值和设定权重，实现了评论数据的情感量化。
S530，根据赋值、权重及权重计算方法，分别得到消极词和积极词的得分，将得分进行汇总得到评论数据的情感值分析结果。例如，基于步骤S520的实施例，对于一条评论数据中同时包括有“这道菜非常好吃”和“这家店的服务比较差”时，其计算方式为1*(1+0.75)+(-1*(1-0.25))＝1，得到该条评论数据的情感评估结果1，即具有积极情感的情感分析结果。
示例性地，如图6所示，提供了一种观点抽取方法的流程示意图，示例性地，在实施例中，观点抽取方法的过程可以包括但不限于步骤S610-S630：
S610，对评论数据进行分词，创建图，以分词得到的词作为节点添加至图；例如，其中分词通过把文本分割成若干个词，在分割时进行分词和词性标注处理，并过滤掉停用词，只保留指定词性的单词，如名词、动词、形容词，得到的词作为节点，例如“这道菜的味道很棒”，通过实施例的方式得到“菜”、“味道”、“棒”等节点，并构建图，其中图用于展示所得到的词及其关系。
S620，通过窗口识别图中每个词的关系，对存在关系的词之间绘制连线；例如，基于步骤S610的实施方式，将“菜”、“味道”、“棒”等节点通过设定大小的窗口进行截取，将每个词语作为候选关键词图的节点，截取的每一段文本中的词语作为相邻的边，对存在关系的词之间绘制连线，构建上述候选关键词图。
S630，根据每个节点的连线数量对词进行排序，得到评论数据的观点提取结果。例如，循环迭代如步骤S620得到的候选关键词图，每个节点的权重初始化化，通过设定的迭代次数达到稳定后，对节点权重进行倒序排序，得到观点抽取方法，其中的观点即为关键词。
S200，响应于可视化请求，确定可视化配置，根据可视化配置得到可视化图表。
其中可视化请求是获取第一分析结果并执行可视化的指令，该指令包括所要可视化的第一分析结果及其对应的可视化配置，例如，商户需要对多项数据进行可视化展示，因此，首先获取可视化配置，确定其所需第一分析结果的数据要求，执行第一分析结果的数据拉取，同时，根据可视化配置获取可视化所要展示的形式，例如雷达图、情感变化图等，创建相应的可视化图表。
示例性地，如图7所示，提供了一种可视化配置的流程示意图，在实施例中，可视化配置的流程可以包括但不限于步骤S710-S730：
S710，读取可视化配置，获取可视化格式、可视化图表类型、可视化时间及可视化结果类型。示例性地，某商户想要查看11月的用户画像图，并且显示方式为雷达图，因此可视化格式为所要可视化的数据表现形式，可视化图表类型为雷达图，可视化时间为11月1日-11月31，可视化结果类型为情感分析结果及语义网络分析结果，其中，数据表现形式可以是百分比值或者柱形图等方式。
S720，调用可视化图表接口根据可视化格式及可视化图表生成可视化图表。示例性地，生成可视化图表可以采用调用可视化控件或者调用可视化接口。
S730，根据可视化时间及可视化结果类型，获取分析结果，通过分析结果对可视化图表进行填充。示例性地，在生成可视化图表之后，根据获取可视化格式、可视化时间及可视化结果类型获取可视化时间所要求的分析结果，参考图8，其中口味3.95、星级3.94、服务3.95、品种4.04及环境3.93即为该步骤所要填充的数据。
参考图8，实施例提供了一种用户画像可视化示意图，商户想要查看11月的用户画像图，并且显示方式为雷达图，因此可视化格式为所要可视化的数据表现形式，可视化图表类型为雷达图，可视化时间为11月1日-11月31所产生的评论数据，可视化结果类型为情感分析结果及语义网络分析结果，参考图8，其提供了一种用户画像的可视化示意图，图中口味、星级、服务、品种及环境为语义网络分析结果，而对应的比值包括口味3.95、星级3.94、服务3.95、品种4.04及环境3.93，这些对应的情感分析结果，在一种实施方式中，可视化的数据表现形式5*积极情感/(积极情感+消极情感)。
S300，通过可视化配置及可视化图表对至少一种第一分析结果进行可视化显示。
其中，第一分析结果可以表示为某商户的评论数据的自然语言分析方法的分析结果；其中，在一个可行的实施方案中，可视化配置及可视化图表对至少一种第一分析结果进行可视化显示包括根据可视化指令，对商户的评论数据的至少一种分析结果进行可视化显示，根据可视化配置及分析结果进行展示，如标签抽取、观点聚类、词云展示、情感变化、情感极性。
示例性地，参考图9，实施例提供了一种标签抽取可视化示意图，其中标签抽取包括根据如上述的情感分析结果和观点抽取，本实施例提供了单个标签抽取和商家评论挖掘的标签可视化示意图，该图按照箭头的导向，完成了标签的可视化流程。图中左侧包括了顾客的评论和商户所选的评论观点挖掘，本实施例通过对单个评论执行如网络分析方法、情感分析方法或观点抽取方法的自然语言处理，得到单个评论标签，对该商户所有的标签进行统计后，得到如右下侧的标签抽取结果，以实现从多个角度对商家的销售情况进行充分挖掘。
示例性地，参考图10，图10示例了一种餐饮数据对比分析方法的流程图，该方法具体包括但不限于步骤S1010-S1020。
S1010，根据商户餐饮类型及预设规则获取至少一个其他商户的第二分析结果，将第一分析结果及第二分析结果通过可视化图表进行可视化对比显示。为了为商家全面了解自己的竞争对手提供有效的帮助，通过商户餐饮类型及预设规则确定与商户存在竞争关系的其他商户，例如，商户餐饮类型为商户的主营菜品为湘菜，基于此餐饮类型查找营菜品为湘菜的其他商户，其中的预设规则可以是地理位置距离、订单数量或用户访问量的至少之一，为了查找合适的竞争对手，本实施例通过地理位置、订单数量及用户访问量执行筛选，例如，查找某市某区内的湘菜馆，且其订单数大于1000，用户访问量大于10000。在一些可行的实施例中， 预设规则是一个灵活配置，进一步的，还可以根据例如每日访问/成交的比值，或者是用户新增数量等。
S1020，对比第一分析结果与第二分析结果，得到对比结果，根据对比结果或第一分析结果生成建议文本，通过可视化图表对建议文本进行可视化显示；其中第二分析结果为其他商户基于第一分析结果的可视化配置及可视化图表得到，进一步地，将所有商户的分析结果，采用相同的配置表进行可视化显示，即可为商户提供可视化的顾客需求和满意度指导。其中，建议文本通过将第一分析结果及第二分析结果通过语义网络分析方法、情感分析方法或观点抽取方法的至少之一得到的相同词和关键词进行情感对比，确定情感区别，根据情感区别自动生成建议或者评估文本，示例性地，对于两个商户，其语义网络分析方法得到“服务”这一词，通过对商家包括有“服务”的评论进行情感分析，分别得到积极评论“服务(617)”和“服务(98)”，通过对比确定积极评论“服务(617)”的商户的服务好于“服务(98)”的商户，对于服务较差的商户，生成如下的建议文本“消费者对服务更加关注，因此提高服务度对增加好评具有一定的作用”，其中建议文本通过固定格式文本加变动文本的方式生成，如上述“服务”为变动文本，而“消费者对__更加关注，因此提高__度对增加好评具有一定的作用”为固定格式文本，在一些实施方式中，建议文本的生成方法不局限于上述的实施方式，还可以采用诸如机器学习的方式，将不同的建议作为机器学习模型的输入，输出具有建议性的建议或者评估结果。
示例性地，参考图11，提供了一种情感指数变化对比可视化的示意图，其示例了两个商户C1和C2在不同的时间段内的情感指数变化，图中的横坐标为时间，纵坐标为情感值。首先图中的商家C1和C2具有相同的餐饮类型，通过可视化配置的选取的时间段和可视化图表，展示了时间段内具有最高情感指数的商户C1和具有最低情感指数的商户C2，即商户C1为上方的情感指数变化的波形，即商户C2为C1下方的情感指数变化的波形。通过该实施例，将不同的商户的顾客的评论数据进行了情感可视化的对比。
示例性地，参考图12，提供了一种竞品标签的可视化对比示意图，图中示例了三个商户D1、D2及D3的不同标签及标签的情感统计，在该实施例中，首选通过语义网络分析方法，确定商户D1、D2及D3各自的语义分析结果，如商户D1的语义分析结果为味道、服务、环境、好吃、新鲜、态度、口味、价格、排队、等、贵、差、便宜、卫生、性价比、物美价廉，然后对语义词对应的评论数据执行情感分析方法，即得到味道896、服务486、环境386、好吃307、新鲜294、态度167、口味117、价格94、排队83、等73、贵71、差62便宜62、卫生46、性价比46、物美价廉13等可视化分析结果。又图，本实施例还示意了根据用户选择积极评论和消极评论，如D2的商户为消极评论，D1和D3包括积极评论和消极评论的综合可视化结果。
示例性地，参考图13，图13示例了一种商户餐饮数据年度可视化报告示意图，图中包括5个页面P1、P2、P3、P4、P5，其中P1为商户餐饮数据2020年度可视化报告的主页面的可视化图，P2为词云展示分析结果的可视化示意图，P3为某一用户的最佳评论(即标签抽取)，P4为热评菜品的排序结果，P5为生成建议或者评估文本。需要说明的是，P1中为通过商户的可视化请求，该请求为年度报告，可以知道，其获取可视化格式、可视化图表类型、可视化时间及可视化结果类型分别，在该实施例中可视化格式为多页面展示，可视化图表类型为“年度报告”，可视化时间为2020.1.1～2020.12.31，可视化结果类型包括如P1～P5所示的综合分析结果，该分析结果通过语义网络分析方法、情感分析方法或观点抽取方法中的多个生成，如P2中的词云展示，其通过语义网络分析方法得到语义词，并难过观点抽取方法进行观点抽取，得到词云分析结果，并可视化于P2页面；又如P5中的通过对多个商家包括有的评论进行情感分析，分别得到“服务”和“菜品”，与其他商家对比后这两个词的情感分析结果较差，生成“服务”和“菜品”的建议或者评估文本。
如图14所示，本申请实施例还提供了一种餐饮数据分析分析装置，该装置包括了评论数 据分析模块1401、可视化图表生成模块1402以及可视化显示模块1403；
其中，评论数据分析模块，用于获取评论数据，通过自然语言分析方法执行分析，得到第一分析结果；可视化图表生成模块，用于根据可视化请求，确定可视化配置，根据可视化配置得到可视化图表；可视化显示模块，用于通过可视化配置及可视化图表对至少一种第一分析结果进行可视化显示。
示例性地，在装置中的评论数据分析模块、可视化图表生成模块以及可视化显示模块的协同下，实施例装置可以实现前述的任意一种餐饮数据分析方法，即获取评论数据，通过自然语言分析方法执行分析；确定可视化配置，根据所述可视化配置得到可视化图表；通过所述可视化配置及所述可视化图表对分析结果进行可视化显示。本申请通过对餐饮商家在售卖餐饮产品时所产生的餐饮数据执行多种自然语言分析，确定评论数据中顾客的偏好、满意度和意见反馈；通过可视化配置及可视化显示，为餐饮商家提供了所需求的可视化显示方式，通过自然语言的综合分析分析可以快速地了解消费者的关注点、需求点，以及商品本身的不足点和店铺服务相关的问题点；为商家提供更精准的顾客需求和满意度指导。
本申请实施例还提供了一种电子设备，该电子设备包括处理器以及存储器；
存储器存储有程序；
处理器执行所述程序实现一种餐饮数据分析方法，其中，所述餐饮数据分析方法包括：获取评论数据，通过自然语言分析方法执行分析，得到第一分析结果；响应于可视化请求，确定可视化配置，根据所述可视化配置得到可视化图表；通过所述可视化配置及所述可视化图表对至少一种所述第一分析结果进行可视化显示。此外；该电子设备具有搭载并运行本申请实施例提供的餐饮数据分析的软件系统的功能，例如，个人计算机(Personal Computer，PC)、手机、智能手机、个人数字助手(Personal Digital Assistant，PDA)、可穿戴设备、掌上电脑PPC(Pocket PC)、平板电脑等。
本申请实施例还提供了一种计算机可读存储介质，所述存储介质存储有程序，所述程序被处理器执行实现一种餐饮数据分析方法，其中，餐饮数据分析方法包括：获取评论数据，通过自然语言分析方法执行分析，得到第一分析结果；响应于可视化请求，确定可视化配置，根据所述可视化配置得到可视化图表；通过所述可视化配置及所述可视化图表对至少一种所述第一分析结果进行可视化显示。此外，所述计算机可读存储介质可以是非易失性，也可以是易失性。
在一些可选择的实施例中，在方框图中提到的功能/操作可以不按照操作示图提到的顺序发生。例如，取决于所涉及的功能/操作，连续示出的两个方框实际上可以被大体上同时地执行或所述方框有时能以相反顺序被执行。此外，在本申请的流程图中所呈现和描述的实施例以示例的方式被提供，目的在于提供对技术更全面的理解。所公开的方法不限于本文所呈现的操作和逻辑流程。可选择的实施例是可预期的，其中各种操作的顺序被改变以及其中被描述为较大操作的一部分的子操作被独立地执行。
本申请实施例还公开了一种计算机程序产品或计算机程序，该计算机程序产品或计算机程序包括计算机指令，该计算机指令存储在计算机可读存储介质中。计算机设备的处理器可以从计算机可读存储介质读取该计算机指令，处理器执行该计算机指令，使得该计算机设备执行一种餐饮数据分析方法，其中，所述餐饮数据分析方法包括：获取评论数据，通过自然语言分析方法执行分析，得到第一分析结果；响应于可视化请求，确定可视化配置，根据可视化配置得到可视化图表；通过所述可视化配置及所述可视化图表对至少一种所述第一分析结果进行可视化显示。
此外，虽然在功能性模块的背景下描述了本申请，但应当理解的是，除非另有相反说明，所述的功能和/或特征中的一个或多个可以被集成在单个物理装置和/或软件模块中，或者一个或多个功能和/或特征可以在单独的物理装置或软件模块中被实现。还可以理解的是，有关每个模块的实际实现的详细讨论对于理解本申请是不必要的。更确切地说，考虑到在本文中公 开的装置中各种功能模块的属性、功能和内部关系的情况下，在工程师的常规技术内将会了解该模块的实际实现。因此，本领域技术人员运用普通技术就能够在无需过度试验的情况下实现在权利要求书中所阐明的本申请。还可以理解的是，所公开的特定概念仅仅是说明性的，并不意在限制本申请的范围，本申请的范围由所附权利要求书及其等同方案的全部范围来决定。
所述功能如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行本申请各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(ROM，Read-Only Memory)、随机存取存储器(RAM，Random Access Memory)、磁碟或者光盘等各种可以存储程序代码的介质。
在流程图中表示或在此以其他方式描述的逻辑和/或步骤，例如，可以被认为是用于实现逻辑功能的可执行指令的定序列表，可以具体实现在任何计算机可读介质中，以供指令执行系统、装置或设备(如基于计算机的系统、包括处理器的系统或其他可以从指令执行系统、装置或设备取指令并执行指令的系统)使用，或结合这些指令执行系统、装置或设备而使用。就本说明书而言，“计算机可读介质”可以是任何可以包含、存储、通信、传播或传输程序以供指令执行系统、装置或设备或结合这些指令执行系统、装置或设备而使用的装置。
计算机可读介质的更具体的示例(非穷尽性列表)包括以下：具有一个或多个布线的电连接部(电子装置)，便携式计算机盘盒(磁装置)，随机存取存储器(RAM)，只读存储器(ROM)，可擦除可编辑只读存储器(EPROM或闪速存储器)，光纤装置，以及便携式光盘只读存储器(CDROM)。另外，计算机可读介质甚至可以是可在其上打印所述程序的纸或其他合适的介质，因为可以例如通过对纸或其他介质进行光学扫描，接着进行编辑、解译或必要时以其他合适方式进行处理来以电子方式获得所述程序，然后将其存储在计算机存储器中。
应当理解，本申请的各部分可以用硬件、软件、固件或它们的组合来实现。在上述实施方式中，多个步骤或方法可以用存储在存储器中且由合适的指令执行系统执行的软件或固件来实现。例如，如果用硬件来实现，和在另一实施方式中一样，可用本领域公知的下列技术中的任一项或他们的组合来实现：具有用于对数据信号实现逻辑功能的逻辑门电路的离散逻辑电路，具有合适的组合逻辑门电路的专用集成电路，可编程门阵列(PGA)，现场可编程门阵列(FPGA)等。
在本说明书的描述中，参考术语“一个实施例”、“一些实施例”、“示例”、“具体示例”、或“一些示例”等的描述意指结合该实施例或示例描述的具体特征、结构、材料或者特点包含于本申请的至少一个实施例或示例中。在本说明书中，对上述术语的示意性表述不一定指的是相同的实施例或示例。而且，描述的具体特征、结构、材料或者特点可以在任何的一个或多个实施例或示例中以合适的方式结合。
尽管已经示出和描述了本申请的实施例，本领域的普通技术人员可以理解：在不脱离本申请的原理和宗旨的情况下可以对这些实施例进行多种变化、修改、替换和变型，本申请的范围由权利要求及其等同物限定。
以上是对本申请的较佳实施进行了具体说明，但本申请并不限于所述实施例，熟悉本领域的技术人员在不违背本申请精神的前提下还可做出种种的等同变形或替换，这些等同的变形或替换均包含在本申请权利要求所限定的范围内。

Claims (20)

  1. 一种餐饮数据分析方法，其中，包括：
    获取评论数据，通过自然语言分析方法执行分析，得到第一分析结果；
    响应于可视化请求，确定可视化配置，根据所述可视化配置得到可视化图表；
    通过所述可视化配置及所述可视化图表对至少一种所述第一分析结果进行可视化显示。
  2. 根据权利要求1所述的餐饮数据分析方法，其中，所述自然语言分析方法包括语义网络分析方法、情感分析方法和观点抽取方法中的至少之一。
  3. 根据权利要求2所述的餐饮数据分析方法，其中，所述语义网络分析方法包括：
    提取所述评论数据中使用频率或出现次数超过预设值的高频词；
    对所述高频词执行无意义词过滤及特征提取，得到所述高频词的特征向量；
    创建共现矩阵，将所述特征向量输入至所述共现矩阵，得到语义网络分析结果。
  4. 根据权利要求2所述的餐饮数据分析方法，其中，所述情感分析方法包括：
    通过情感词典对所述评论数据执行情感词分析，得到消极词、积极词、程度副词及程度符号；
    对所述消极词和所述积极词分别赋值，并为所述程度副词及所述程度符号分配权重和权重计算方法；
    根据所述赋值、所述权重及所述权重计算方法，分别得到所述消极词和所述积极词的得分，将所述得分进行汇总得到所述评论数据的情感值分析结果。
  5. 根据权利要求2所述的餐饮数据分析方法，其中，所述观点抽取方法包括：
    对所述评论数据进行分词并创建图，以分词得到的词作为节点添加至所述图；
    通过窗口识别所述图中每个词的关系，对存在关系的所述词之间绘制连线；
    根据每个所述节点的连线数量对所述词进行排序，得到所述评论数据的观点提取结果。
  6. 根据权利要求1所述的餐饮数据分析方法，其中，所述响应于可视化请求，确定可视化配置，根据所述可视化配置得到可视化图表包括：
    读取所述可视化配置，获取可视化格式、可视化图表类型、可视化时间及可视化结果类型；
    调用可视化图表接口根据所述可视化格式及所述可视化图表生成可视化图表；
    根据所述可视化时间及所述可视化结果类型，获取所述分析结果，通过所述分析结果对所述可视化图表进行填充。
  7. 根据权利要求1所述的餐饮数据分析方法，其中，所述通过所述可视化配置及所述可视化图表对至少一种所述第一分析结果进行可视化显示，还包括：
    根据所述可视化请求，确定商户餐饮类型；
    根据所述商户餐饮类型及预设规则获取至少一个其他商户的第二分析结果，将所述第一分析结果及所述第二分析结果通过所述可视化图表进行可视化对比显示；
    以及，对比所述第一分析结果与所述第二分析结果，得到对比结果，根据所述对比结果或第一分析结果生成建议文本，通过所述可视化图表对所述建议文本进行可视化显示；
    所述预设规则包括地理位置距离、订单数量和用户访问量中的至少之一。
  8. 一种餐饮数据分析装置，其中，包括：
    评论数据分析模块，用于获取评论数据，通过自然语言分析方法执行分析，得到第一分析结果；
    可视化图表生成模块，用于根据可视化请求，确定可视化配置，根据所述可视化配置得到可视化图表；
    可视化显示模块，用于通过所述可视化配置及所述可视化图表对至少一种所述第一分析结果进行可视化显示。
  9. 一种电子设备，其中，包括处理器以及存储器；
    所述存储器用于存储程序；
    所述处理器执行所述程序实现一种餐饮数据分析方法，其中，所述餐饮数据分析方法包括：
    获取评论数据，通过自然语言分析方法执行分析，得到第一分析结果；
    响应于可视化请求，确定可视化配置，根据所述可视化配置得到可视化图表；
    通过所述可视化配置及所述可视化图表对至少一种所述第一分析结果进行可视化显示。
  10. 根据权利要求9所述的电子设备，其中，所述自然语言分析方法包括语义网络分析方法、情感分析方法和观点抽取方法中的至少之一。
  11. 根据权利要求10所述的电子设备，其中，所述语义网络分析方法包括：
    提取所述评论数据中使用频率或出现次数超过预设值的高频词；
    对所述高频词执行无意义词过滤及特征提取，得到所述高频词的特征向量；
    创建共现矩阵，将所述特征向量输入至所述共现矩阵，得到语义网络分析结果。
  12. 根据权利要求10所述的电子设备，其中，所述情感分析方法包括：
    通过情感词典对所述评论数据执行情感词分析，得到消极词、积极词、程度副词及程度符号；
    对所述消极词和所述积极词分别赋值，并为所述程度副词及所述程度符号分配权重和权重计算方法；
    根据所述赋值、所述权重及所述权重计算方法，分别得到所述消极词和所述积极词的得分，将所述得分进行汇总得到所述评论数据的情感值分析结果。
  13. 根据权利要求10所述的电子设备，其中，所述观点抽取方法包括：
    对所述评论数据进行分词并创建图，以分词得到的词作为节点添加至所述图；
    通过窗口识别所述图中每个词的关系，对存在关系的所述词之间绘制连线；
    根据每个所述节点的连线数量对所述词进行排序，得到所述评论数据的观点提取结果。
  14. 根据权利要求9所述的电子设备，其中，所述通过所述可视化配置及所述可视化图表对至少一种所述第一分析结果进行可视化显示，还包括：
    根据所述可视化请求，确定商户餐饮类型；
    根据所述商户餐饮类型及预设规则获取至少一个其他商户的第二分析结果，将所述第一分析结果及所述第二分析结果通过所述可视化图表进行可视化对比显示；
    以及，对比所述第一分析结果与所述第二分析结果，得到对比结果，根据所述对比结果或第一分析结果生成建议文本，通过所述可视化图表对所述建议文本进行可视化显示；
    所述预设规则包括地理位置距离、订单数量和用户访问量中的至少之一。
  15. 一种计算机可读存储介质，其中，所述存储介质存储有程序，所述程序被处理器执行实现一种餐饮数据分析方法，其中，所述餐饮数据分析方法包括：
    获取评论数据，通过自然语言分析方法执行分析，得到第一分析结果；
    响应于可视化请求，确定可视化配置，根据所述可视化配置得到可视化图表；
    通过所述可视化配置及所述可视化图表对至少一种所述第一分析结果进行可视化显示。
  16. 根据权利要求15所述的计算机可读存储介质，其中，所述自然语言分析方法包括语义网络分析方法、情感分析方法和观点抽取方法中的至少之一。
  17. 根据权利要求16所述的计算机可读存储介质，其中，所述语义网络分析方法包括：
    提取所述评论数据中使用频率或出现次数超过预设值的高频词；
    对所述高频词执行无意义词过滤及特征提取，得到所述高频词的特征向量；
    创建共现矩阵，将所述特征向量输入至所述共现矩阵，得到语义网络分析结果。
  18. 根据权利要求16所述的计算机可读存储介质，其中，所述情感分析方法包括：
    通过情感词典对所述评论数据执行情感词分析，得到消极词、积极词、程度副词及程度符号；
    对所述消极词和所述积极词分别赋值，并为所述程度副词及所述程度符号分配权重和权 重计算方法；
    根据所述赋值、所述权重及所述权重计算方法，分别得到所述消极词和所述积极词的得分，将所述得分进行汇总得到所述评论数据的情感值分析结果。
  19. 根据权利要求16所述的计算机可读存储介质，其中，所述观点抽取方法包括：
    对所述评论数据进行分词并创建图，以分词得到的词作为节点添加至所述图；
    通过窗口识别所述图中每个词的关系，对存在关系的所述词之间绘制连线；
    根据每个所述节点的连线数量对所述词进行排序，得到所述评论数据的观点提取结果。
  20. 根据权利要求15所述的计算机可读存储介质，其中，所述通过所述可视化配置及所述可视化图表对至少一种所述第一分析结果进行可视化显示，还包括：
    根据所述可视化请求，确定商户餐饮类型；
    根据所述商户餐饮类型及预设规则获取至少一个其他商户的第二分析结果，将所述第一分析结果及所述第二分析结果通过所述可视化图表进行可视化对比显示；
    以及，对比所述第一分析结果与所述第二分析结果，得到对比结果，根据所述对比结果或第一分析结果生成建议文本，通过所述可视化图表对所述建议文本进行可视化显示；
    所述预设规则包括地理位置距离、订单数量和用户访问量中的至少之一。
PCT/CN2022/090739 2022-02-22 2022-04-29 餐饮数据分析方法、装置、电子设备及存储介质 WO2023159766A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210163511.XA CN114519100A (zh) 2022-02-22 2022-02-22 餐饮数据分析方法、装置、电子设备及存储介质
CN202210163511.X 2022-02-22

Publications (1)

Publication Number Publication Date
WO2023159766A1 true WO2023159766A1 (zh) 2023-08-31

Family

ID=81600054

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/090739 WO2023159766A1 (zh) 2022-02-22 2022-04-29 餐饮数据分析方法、装置、电子设备及存储介质

Country Status (2)

Country Link
CN (1) CN114519100A (zh)
WO (1) WO2023159766A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115392199B (zh) * 2022-08-22 2023-08-04 再惠(上海)网络科技有限公司 评价分析和报告生成的方法、装置、电子设备及存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150302475A1 (en) * 2013-11-20 2015-10-22 Sujit Thomas Zachariah Managing, publishing and tracking content for virtual store display windows
CN108536671A (zh) * 2018-03-07 2018-09-14 世纪龙信息网络有限责任公司 文本数据的情感指数识别方法和系统
CN109003113A (zh) * 2018-05-30 2018-12-14 浙江口碑网络技术有限公司 评价数据处理和展示的方法及装置，电子设备及存储设备
CN109697657A (zh) * 2018-12-27 2019-04-30 厦门快商通信息技术有限公司 一种餐饮推荐方法、服务器及存储介质
CN111523923A (zh) * 2020-04-06 2020-08-11 北京三快在线科技有限公司 商家评论管理系统、方法、服务器及存储介质

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150302475A1 (en) * 2013-11-20 2015-10-22 Sujit Thomas Zachariah Managing, publishing and tracking content for virtual store display windows
CN108536671A (zh) * 2018-03-07 2018-09-14 世纪龙信息网络有限责任公司 文本数据的情感指数识别方法和系统
CN109003113A (zh) * 2018-05-30 2018-12-14 浙江口碑网络技术有限公司 评价数据处理和展示的方法及装置，电子设备及存储设备
CN109697657A (zh) * 2018-12-27 2019-04-30 厦门快商通信息技术有限公司 一种餐饮推荐方法、服务器及存储介质
CN111523923A (zh) * 2020-04-06 2020-08-11 北京三快在线科技有限公司 商家评论管理系统、方法、服务器及存储介质

Also Published As

Publication number Publication date
CN114519100A (zh) 2022-05-20

Similar Documents

Publication Publication Date Title
Kumar et al. Sentiment analysis of multimodal twitter data
Wu et al. OpinionSeer: interactive visualization of hotel customer feedback
Tse et al. Insight from the horsemeat scandal: Exploring the consumers’ opinion of tweets toward Tesco
US9672555B1 (en) Extracting quotes from customer reviews
US20150186790A1 (en) Systems and Methods for Automatic Understanding of Consumer Evaluations of Product Attributes from Consumer-Generated Reviews
KR20190108838A (ko) 미술 작품 추천 큐레이션 방법 및 시스템
TW201423450A (zh) 基於電子資訊的關鍵字提取的資訊推送、搜尋方法及裝置
CN111523923A (zh) 商家评论管理系统、方法、服务器及存储介质
US20200192921A1 (en) Suggesting text in an electronic document
Petz et al. On text preprocessing for opinion mining outside of laboratory environments
JP2018169986A (ja) 企業情報提供システムおよびプログラム
Krestel et al. Diversifying customer review rankings
Hu et al. A novel approach to rate and summarize online reviews according to user-specified aspects
Nasim et al. ABSA toolkit: An open source tool for aspect based sentiment analysis
Tayal et al. Personalized ranking of products using aspect-based sentiment analysis and Plithogenic sets
Zhang et al. Using neutral sentiment reviews to improve customer requirement identification and product design strategies
CN113434755A (zh) 页面的生成方法、装置、电子设备及存储介质
Nareshkumar et al. Interactive deep neural network for aspect-level sentiment analysis
WO2023159766A1 (zh) 餐饮数据分析方法、装置、电子设备及存储介质
CN116882414B (zh) 基于大规模语言模型的评语自动生成方法及相关装置
Nguyen et al. Analyzing Customer Experience in Hotel Services Using Topic Modeling.
WO2023231288A1 (zh) 搜索结果的排序方法、装置、电子设备和存储介质
Gurin Methods for Automatic Sentiment Detection
Jiang et al. Leveraging aspect phrase embeddings for cross-domain review rating prediction
KR20220118703A (ko) 머신러닝 기반 온라인 쇼핑 리뷰 감정 예측 시스템 및 방법

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22928037

Country of ref document: EP

Kind code of ref document: A1