User demand analysis method and device, equipment, storage medium and program product
The process of community chat data through TF-IDF and GN algorithms solves the problem of low accuracy in user demand analysis, achieves more accurate user demand analysis and product optimization, and improves after-sales service and marketing effects.
Patent Information
- Application Number
- CN202411042901.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-31
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2044-07-31
AI Technical Summary
In the prior art, the accuracy of user demand analysis is low, affected by anomalies and noise, and the kmeans algorithm is sensitive to the initial random seed points, resulting in unstable results.
The TF-IDF text word segmentation algorithm and the GN community discovery algorithm are used to process community chat data, build a word network, automatically iterate and classify, reduce the impact of noise, and improve clustering accuracy.
It improves the accuracy of user demand analysis, enhances the accuracy and relevance of clustering, helps the product department optimize product design, and improves after-sales service capabilities and precise marketing effects.
Smart Images

Figure CN119067108A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of vehicle technology, and in particular to a user demand analysis method, device, equipment, storage medium and program product. Background Art
[0002] As competition among auto companies intensifies, the relationship between auto companies and users becomes deeper, and users have more channels to express themselves. Auto companies have a large amount of user text data. As the amount of text data grows, manual summarization will increase the workload. Current products serve users throughout their life cycle and provide comprehensive services to users, discovering user pain points and opportunities from their voices.
[0003] The main method for analyzing user needs is word clustering, among which text segmentation + kmeans is widely used. However, this method has several problems: 1. Clustering based on the kmeans algorithm will be affected by outliers or noise. For example, "I like traveling" and "I often go hiking" actually express the meaning of traveling, but kmeans will not treat these two sentences as the same words; 2. The kmeans algorithm requires an initial random seed point to start optimization. This random seed point is too important. Different random seed points will produce completely different results. Therefore, the above method is not accurate enough for analyzing user needs. Summary of the invention
[0004] The present application provides a user demand analysis method, device, equipment, storage medium and program product to solve the problem that the related technology has low accuracy in analyzing user needs.
[0005] The first aspect of the present application provides a user demand analysis method, comprising the following steps: obtaining user data and community chat data of private domain users of a target automobile company; inputting the community chat data into a target model, wherein the target model processes the community chat data using a TF-IDF text segmentation algorithm and a GN community discovery algorithm in sequence, and outputs chat topics; analyzing the needs of private domain users based on the chat topics and user data.
[0006] Optionally, in one embodiment of the present application, after obtaining the user data and community chat data of the private domain users of the target automobile company, it includes: performing data processing on the user data and community chat data to obtain processed user data information and community chat data; and inputting the processed community chat data into the target model.
[0007] Optionally, in one embodiment of the present application, the community chat data is processed using a TF-IDF algorithm, including: obtaining each piece of community chat data and all community chat data of all private domain users; calculating the frequency of each word in each piece of community chat data; calculating the inverse document frequency of each word in all community chat data; calculating the TF-IDF value of each word based on the frequency and the inverse document frequency, and determining multiple keywords based on the TF-IDF value.
[0008] Optionally, in one embodiment of the present application, the community chat data is processed by the GN algorithm, including: establishing at least one edge based on the frequency of occurrence of multiple keywords in each community chat, and using multiple keywords as nodes; constructing a word network based on the nodes and at least one edge, calculating the number of boundaries of each edge in the word network, and deleting the edge with the largest number of boundaries, recalculating the number of boundaries in the word network, and stopping the calculation until at least one node in the word network meets a preset condition, and dividing keywords with similar meanings into corresponding topic clusters.
[0009] Optionally, in one embodiment of the present application, the user data includes: user basic information and user vehicle information, and the data processing includes at least one of data cleaning, data integration, data transformation and data reduction.
[0010] Optionally, in one embodiment of the present application, determining multiple keywords according to TF-IDF values includes: arranging all words in descending order according to TF-IDF values; and selecting words with TF-IDF values within a target range as keywords.
[0011] The second aspect of the present application provides a user demand analysis device, including: an acquisition module, used to obtain user data and community chat data of private domain users under a target automobile enterprise; an input module, used to input the community chat data into a target model, wherein the target model processes the community chat data using a TF-IDF text segmentation algorithm and a GN community discovery algorithm in sequence, and outputs a chat topic; an analysis module, used to analyze the needs of private domain users based on the chat topic and user data.
[0012] Optionally, in one embodiment of the present application, it also includes: a processing module, which is used to process the user data and community chat data of private domain users under the target automobile company after acquiring the user data and community chat data to obtain processed user data information and community chat data; and input the processed community chat data into the target model.
[0013] Optionally, in one embodiment of the present application, the community chat data is processed using a TF-IDF algorithm, including: obtaining each piece of community chat data and all community chat data of all private domain users; calculating the frequency of each word in each piece of community chat data; calculating the inverse document frequency of each word in all community chat data; calculating the TF-IDF value of each word based on the frequency and the inverse document frequency, and determining multiple keywords based on the TF-IDF value.
[0014] Optionally, in one embodiment of the present application, the community chat data is processed by the GN algorithm, including: establishing at least one edge based on the frequency of occurrence of multiple keywords in each community chat, and using multiple keywords as nodes; constructing a word network based on the nodes and at least one edge, calculating the number of boundaries of each edge in the word network, and deleting the edge with the largest number of boundaries, recalculating the number of boundaries in the word network, and stopping the calculation until at least one node in the word network meets a preset condition, and dividing keywords with similar meanings into corresponding topic clusters.
[0015] Optionally, in one embodiment of the present application, the user data includes: user basic information and user vehicle information, and the data processing includes at least one of data cleaning, data integration, data transformation and data reduction.
[0016] Optionally, in one embodiment of the present application, determining multiple keywords according to TF-IDF values includes: arranging all words in descending order according to TF-IDF values; and selecting words with TF-IDF values within a target range as keywords.
[0017] The third aspect of the present application provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to perform a user demand analysis method as described in the above embodiment.
[0018] The fourth aspect of the present application provides a computer-readable storage medium on which a computer program or instruction is stored. The computer program or instruction is executed by a processor to perform the user demand analysis method as described in the above embodiment.
[0019] The fifth aspect of the present application provides a computer program product, including a computer program or instructions. When the computer program or instructions are executed, the user demand analysis method described in the above embodiment is implemented.
[0020] Therefore, this application has at least the following beneficial effects:
[0021] The embodiment of the present application can obtain the chat topic of the user based on the recognition and analysis method constructed based on the algorithm model according to the data information of the private domain user, chat data information, etc. The target model includes the TF-IDF algorithm and the GN algorithm. The GN algorithm is not affected by noise, and can also construct the correlation and inheritance between the groups to improve the accuracy of clustering, thereby improving the accuracy of user demand analysis. Subsequently, the user's needs are determined based on the chat topic and user data, and the user needs are analyzed in depth and comprehensively, which effectively helps the product department to optimize the product design. At the same time, through a comprehensive analysis of user needs, the corresponding car companies can improve their after-sales service capabilities and carry out precision marketing. Therefore, the technical problems such as the low accuracy of the analysis of user needs in the relevant technology are solved.
[0022] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS
[0023] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:
[0024] Figure 1 A flowchart of a user demand analysis method provided according to an embodiment of the present application;
[0025] Figure 2 A flowchart of data processing according to an embodiment of the present application;
[0026] Figure 3 A schematic diagram of data cleaning according to an embodiment of the present application;
[0027] Figure 4 A schematic diagram of data conversion according to an embodiment of the present application;
[0028] Figure 5 A schematic diagram of a target model constructed according to an embodiment of the present application;
[0029] Figure 6 This is an example diagram of a user demand analysis device provided according to an embodiment of the present application;
[0030] Figure 7 It is a schematic diagram of the structure of an electronic device provided according to an embodiment of the present application. DETAILED DESCRIPTION
[0031] Embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.
[0032] The user demand analysis method, device, equipment, storage medium and program product of the embodiment of the present application are described below with reference to the accompanying drawings. In view of the problem mentioned in the background technology center that the current use of text segmentation and kmeans for user demand analysis is not accurate enough in the analysis structure, the present application provides a user demand analysis method, in which the user's chat topic can be obtained based on the data information such as private domain user data and chat data information, and the recognition analysis method based on the algorithm model construction. The target model includes the TF-IDF algorithm and the GN algorithm. The GN algorithm is not affected by noise, and can also construct the correlation and inheritance between groups to improve the accuracy of clustering, thereby improving the accuracy of user demand analysis, and subsequently determining the user's needs based on the chat topic and user data. As a result, the problem of low accuracy of related technologies in analyzing user needs is solved.
[0033] Specifically, Figure 1 A flowchart of a user demand analysis method provided in an embodiment of the present application.
[0034] like Figure 1 As shown, the user demand analysis method includes the following steps:
[0035] In step S101, the user data and community chat data of the private domain users of the target automobile enterprise are obtained.
[0036] Among them, user data includes: user basic information and user vehicle information, both of which can be obtained through the database; user basic information data includes name, gender, age, native place, education, major, graduation school, marital status, home address, years of work, position, etc.; user vehicle information includes: vehicle model, vehicle price, purchase time, insurance, etc. Examples of user data and community chat data are shown in Table 1, which is a data table of different dimensions of private domain users.
[0037] Table 1
[0038]
[0039]
[0040] It can be understood that the embodiments of the present application can obtain user data and community chat data of private domain users of the target car company, so as to subsequently analyze user needs based on the data.
[0041] In an embodiment of the present application, after obtaining the user data and community chat data of the private domain users of the target automobile company, it includes: performing data processing on the user data and community chat data to obtain processed user data information and community chat data; and inputting the processed community chat data into the target model.
[0042] Among them, data processing includes at least one of data cleaning, data integration, data transformation and data reduction.
[0043] It is understandable that the embodiments of the present application can perform data processing on user data and community chat records to obtain processed user data information and community chat data, and input the processed community chat data into the target model.
[0044] The data processing process is as follows Figure 2 As shown, including:
[0045] 1. Data cleaning: mainly to delete irrelevant and duplicate data in the original data set of private domain users, smooth the noise data, filter out data irrelevant to the mining topic, and handle missing values and outliers. Specific commonly used methods include deletion, replacement, interpolation, etc.
[0046] 2. Data integration: Integrate the collected, interrelated, distributed, heterogeneous private domain user data sources together, so that users can access these data sources in a transparent manner. Specific integration methods include pattern integration, data replication, and comprehensive integration.
[0047] 3. Data transformation: the process of changing various types of private domain user data from one form of representation to another. Common transformation methods include logarithmic transformation, square root transformation, and reciprocal transformation.
[0048] 4. Data reduction: All kinds of private domain user data collected are reduced while maintaining the integrity of the data set as much as possible to ensure more effective data set mining. Feature reduction, sample reduction, and feature value reduction are mainly used.
[0049] Taking the user data and community chat data in Table 1 as an example, the data is processed. It can be seen from the data that the data of the last "Li XX" user is repeatedly collected. When doing data cleaning, the deletion method can be used to remove the data with repeated collection. The process is as follows: Figure 3 shown.
[0050] Through data transformation, the numerical indicator "age" in the data is converted into a classification indicator according to the rule "18-30 years old - youth, 30-50 years old - middle-aged, over 50 years old - elderly". The results are as follows Figure 4 Through the data processing work in the above steps, the data is more standardized, which is convenient for the next step of data analysis.
[0051] In step S102, the community chat data is input into the target model, wherein the target model processes the community chat data using the TF-IDF text segmentation algorithm and the GN community discovery algorithm in sequence, and outputs the chat topic.
[0052] It can be understood that the embodiment of the present application can input community chat data into the target model, and the target model processes the community chat data using the TF-IDF algorithm and the GN algorithm in sequence, and outputs the chat topic, that is, extracts a topic from the community chat data.
[0053] In an embodiment of the present application, the community chat data is processed by the TF-IDF algorithm, including: obtaining each piece of community chat data and all community chat data of all private domain users; calculating the frequency of each word in each piece of community chat data; calculating the inverse document frequency of each word in all community chat data; calculating the TF-IDF value of each word based on the frequency and the inverse document frequency, and determining multiple keywords based on the TF-IDF value.
[0054] It can be understood that the embodiments of the present application can obtain each community chat record and all set community chat data of all private domain users, calculate the frequency of each word in each community chat data, and calculate the inverse document frequency of each word in all community chat data, and then calculate the TF-IDF value of each word according to the frequency and inverse document frequency, and determine multiple keywords according to the TF-IDF value.
[0055] In an embodiment of the present application, determining multiple keywords according to TF-IDF values includes: arranging all words in descending order according to TF-IDF values; and selecting words with TF-IDF values within a target range as keywords.
[0056] The target range may be the first N words.
[0057] It is understandable that the embodiment of the present application can arrange all words in descending order according to the TF-IDF value, and select words with TF-IDF values within the target range as keywords, such as selecting the top N words.
[0058] In an embodiment of the present application, the community chat data is processed by the GN algorithm, including: establishing at least one edge based on the frequency of appearance of multiple keywords in each community chat, and using multiple keywords as nodes; constructing a word network based on the nodes and at least one edge, calculating the number of boundaries of each edge in the word network, deleting the edge with the largest number of boundaries, recalculating the number of boundaries in the word network, and stopping the calculation until at least one node in the word network meets the preset conditions, and dividing the keywords with similar meanings into corresponding topic clusters.
[0059] It can be understood that the embodiments of the present application can cluster the keywords obtained in the above embodiments, that is, merge similar words and divide the keywords with similar meanings into corresponding topic clusters to determine the corresponding chat topics. For example, the words "car computer" and "car computer screen" both refer to the cockpit car computer, which are similar words. The GN algorithm clustering process is: establish at least one edge according to the frequency of appearance of multiple keywords in each community chat, and use multiple keywords as nodes; construct a word network according to the nodes and at least one edge, calculate the number of boundaries of each edge in the word network, delete the edge with the largest number of boundaries, recalculate the number of boundaries in the word network, and stop the calculation until at least one node in the word network meets the preset conditions, and divide the keywords with similar meanings into corresponding topic clusters.
[0060] It should be noted that the GN algorithm of this application solves the problem that the kmeans clustering method is sensitive to noise and needs to select a suitable random seed, resulting in poor final word clustering and unclear grouping. The GN algorithm can automatically iterate the classification and is not affected by noise. The GN algorithm can also construct the correlation and inheritance between groups.
[0061] Specifically, the target model (i.e. Figure 5 The construction of the recognition model in Figure 5 As shown, it includes: text segmentation and word clustering. The final output result is shown in Table 2. Table 2 is an example table for determining chat topics from chat records.
[0062] Table 2
[0063] Tags (chat topics) Original conversation (chat log) maintainance How often does this car need maintenance? Skylight What do you think of the sunroof on this car? Voice Function How loud is the voice prompt in your car? engine The engine response is pretty fast.
[0064] 1. TF-IDF algorithm used in text segmentation:
[0065] TF (Term Frequency) is simply the frequency at which keywords appear on a web page.
[0066] IDF (Inverse Document Frequency) is simply a data set in which the keyword appears in all documents.
[0067] 1. Calculate word frequency:
[0068]
[0069] 2. Calculate the inverse document frequency:
[0070]
[0071] 3. Calculate TF-IDF:
[0072] TF-IDF = Term Frequency (TF) × Inverse Document Frequency (IDF);
[0073] TF-IDF is directly proportional to the number of times a word appears in a document and inversely proportional to the number of times the word appears in the entire language. Therefore, the algorithm for automatically extracting keywords is to calculate the TF-IDF value of each word in the document, then sort them in descending order, and select the first few words.
[0074] II. Merging similar words - word clustering.
[0075] Classify similar words for the already separated words. For example, the words "car infotainment system" and "car infotainment screen" both refer to the in-vehicle infotainment system in the cockpit, which are similar words. The method used for word clustering is the Girvan–Newman (GN) algorithm:
[0076] 1. Calculate the betweenness centrality of each edge;
[0077] 2. Delete the edge with the highest betweenness centrality;
[0078] 3. Recalculate the betweenness centrality of the remaining edges in the network;
[0079] 4. Repeat steps 3 and 4 until any vertex in the network forms a community.
[0080] In step S103, analyze the needs of private domain users based on the chat topic and user data.
[0081] It can be understood that the embodiments of the present application can analyze the needs of private domain users based on the chat topic and user data, so that automobile enterprises can obtain the needs of private domain users, serve product improvement, discover new marketing points, assist the operation side in formulating operation strategies, improve after-sales service capabilities, and product optimization points.
[0082] In addition, the analysis of the content of private domain users of automobile enterprises based on big data model algorithms can effectively realize the analysis and evaluation of the demands and needs of private domain users, deeply explore the potential needs of private domain users, and provide comprehensive data support for the product department and marketing companies. The present application is different from traditional marketing methods and adopts a digital marketing method, that is, uses big data and algorithm models to analyze the needs of users, and then conducts marketing according to the needs of users. This method can more objectively and efficiently realize the analysis and exploration of user needs, and assist the product side in optimizing product design and marketing personnel in accurate marketing.
[0083] In summary, the present application is based on big data technology and algorithm models to construct a scientific and feasible method for evaluating and analyzing the needs of private domain users. This method is aimed at the problem feedback, demands, and needs analysis of private domain users for existing cars, designs a series of processes from data collection, model construction to result output, and realizes a relatively comprehensive analysis and evaluation of user needs and feedback. The specific solutions are as follows.
[0084] A user demand analysis method based on a big data algorithm model is constructed, following the design principles of scientificity, feasibility, and practicality, and implementing user demand analysis according to the steps of "data collection, data processing, model construction, and model result output."
[0085] Word extraction is performed based on complex network analysis. According to the nature of the problem and the overall goal to be achieved, feature selection technology is used to select suitable candidate words from the chat records of private domain users in the community. Complex network analysis technology is used to use candidate words as nodes and collinearity as edges to link them, forming a huge word network. The GN algorithm is used to cut the word network, and the network eventually becomes word clusters with extremely high internal correlation. The keywords and demands in the user chat records are obtained, and then classified and attributed to product problem feedback, optimization feedback and potential marketing leads.
[0086] The private domain user analysis method based on the big data algorithm model has well verified the feasibility of the analysis method in terms of application through the extraction of private domain user-related product issues and marketing cases. The construction of this method will contribute to the further development and further promote the improvement of product optimization capabilities and marketing accuracy.
[0087] According to the user demand analysis method proposed in the embodiment of the present application, the user's chat topic can be obtained based on the recognition analysis method constructed based on the algorithm model according to the data information such as private domain users' data and chat data information. The target model includes the TF-IDF algorithm and the GN algorithm. The GN algorithm is not affected by noise and can also construct the correlation and inheritance between groups to improve the accuracy of clustering, thereby improving the accuracy of user demand analysis. Subsequently, the user's needs are determined based on the chat topic and user data, and the user needs are analyzed in depth and comprehensively, which effectively helps the product department to optimize the product design. At the same time, through a comprehensive analysis of user needs, the corresponding car companies can improve their after-sales service capabilities and carry out precision marketing.
[0088] Next, the user demand analysis device proposed according to the embodiment of the present application is described with reference to the accompanying drawings.
[0089] Figure 6 It is a block diagram of a user demand analysis device according to an embodiment of the present application.
[0090] like Figure 6 As shown, the user demand analysis device 10 includes: an acquisition module 100, an input module 200 and an analysis module 300.
[0091] Among them, the acquisition module 100 is used to obtain the user data and community chat data of the private domain users of the target car company; the input module 200 is used to input the community chat data into the target model, wherein the target model processes the community chat data with the TF-IDF text segmentation algorithm and the GN community discovery algorithm in turn, and outputs the chat topic; the analysis module 300 is used to analyze the needs of private domain users based on the chat topic and user data.
[0092] In the embodiment of the present application, the device 10 of the embodiment of the present application further includes: a processing module.
[0093] Among them, the processing module is used to process the user data and community chat data of the private domain users of the target automobile enterprise after obtaining them, so as to obtain processed user data information and community chat data; and input the processed community chat data into the target model.
[0094] In an embodiment of the present application, the community chat data is processed by the TF-IDF algorithm, including: obtaining each piece of community chat data and all community chat data of all private domain users; calculating the frequency of each word in each piece of community chat data; calculating the inverse document frequency of each word in all community chat data; calculating the TF-IDF value of each word based on the frequency and the inverse document frequency, and determining multiple keywords based on the TF-IDF value.
[0095] In an embodiment of the present application, the community chat data is processed by the GN algorithm, including: establishing at least one edge based on the frequency of appearance of multiple keywords in each community chat, and using multiple keywords as nodes; constructing a word network based on the nodes and at least one edge, calculating the number of boundaries of each edge in the word network, deleting the edge with the largest number of boundaries, recalculating the number of boundaries in the word network, and stopping the calculation until at least one node in the word network meets the preset conditions, and dividing the keywords with similar meanings into corresponding topic clusters.
[0096] In an embodiment of the present application, user data includes: user basic information and user vehicle information, and data processing includes at least one of data cleaning, data integration, data transformation and data reduction.
[0097] In an embodiment of the present application, determining multiple keywords according to TF-IDF values includes: arranging all words in descending order according to TF-IDF values; and selecting words with TF-IDF values within a target range as keywords.
[0098] It should be noted that the above explanation of the embodiment of the user demand analysis method is also applicable to the user demand analysis device of this embodiment, and will not be repeated here.
[0099] According to the user demand analysis device proposed in the embodiment of the present application, the user's chat topic can be obtained based on the recognition analysis method constructed based on the algorithm model according to the data information such as private domain users' data and chat data information. The target model includes the TF-IDF algorithm and the GN algorithm. The GN algorithm is not affected by noise and can also construct the correlation and inheritance between groups to improve the accuracy of clustering, thereby improving the accuracy of user demand analysis. Subsequently, the user's needs are determined based on the chat topic and user data, and the user needs are analyzed in depth and comprehensively, which effectively helps the product department to optimize the product design. At the same time, through a comprehensive analysis of user needs, the corresponding car companies can improve their after-sales service capabilities and carry out precision marketing.
[0100] Figure 7 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. The electronic device may include:
[0101] A memory 701 , a processor 702 , and a computer program stored in the memory 701 and executable on the processor 702 .
[0102] When the processor 702 executes the program, the user demand analysis method provided in the above embodiment is implemented.
[0103] Furthermore, the electronic device further comprises:
[0104] The communication interface 703 is used for communication between the memory 701 and the processor 702 .
[0105] The memory 701 is used to store computer programs that can be executed on the processor 702 .
[0106] The memory 701 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.
[0107] If the memory 701, the processor 702 and the communication interface 703 are implemented independently, the communication interface 703, the memory 701 and the processor 702 can be connected to each other through a bus and communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 7Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.
[0108] Optionally, in a specific implementation, if the memory 701, the processor 702 and the communication interface 703 are integrated on a chip, the memory 701, the processor 702 and the communication interface 703 can communicate with each other through an internal interface.
[0109] The processor 702 may be a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.
[0110] An embodiment of the present application also provides a computer-readable storage medium on which a computer program or instruction is stored. When the computer program or instruction is executed by a processor, the user demand analysis method as described above is implemented.
[0111] The embodiment of the present application also provides a computer program product, including a computer program or instructions, which, when executed, implements the above user demand analysis method.
[0112] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or N embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.
[0113] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise clearly and specifically defined.
[0114] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or N executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.
[0115] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above embodiment, the N steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one or a combination of multiple of the following technologies known in the art: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.
[0116] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.
Claims
1. A user demand analysis method, characterized in that: The following steps are involved: Obtain user data and community chat data of private users of the target car company; Inputting the community chat data into the target model, wherein the target model processes the community chat data using the TF-IDF text segmentation algorithm and the GN community discovery algorithm in sequence, and outputs the chat topic; The needs of the private domain user are analyzed according to the chat topic and the user data.
2. The user demand analysis method according to claim 1, characterized in that: After obtaining the user data and community chat data of the private domain users of the target car company, including: Performing data processing on the user data and the community chat data to obtain processed user data information and community chat data; The processed community chat data is input into the target model.
3. The user demand analysis method according to claim 1, characterized in that: The processing of the community chat data by the TF-IDF algorithm includes: Get each social chat data and all social chat data of all private domain users; Calculate the frequency of each word in each piece of community chat data appearing in each piece of community chat data; Calculate the inverse document frequency of each word in all the community chat data; The TF-IDF value of each word is calculated according to the frequency and the inverse document frequency, and a plurality of keywords are determined according to the TF-IDF value.
4. The user demand analysis method according to claim 3, characterized in that: The processing of the community chat data by GN algorithm includes: Establish at least one edge according to the frequencies of occurrence of the multiple keywords in each community chat, and use the multiple keywords as nodes; A word network is constructed according to the nodes and the at least one edge, the number of boundaries of each edge in the word network is calculated, and the edge with the largest number of boundaries is deleted, and the number of boundaries in the word network is recalculated until at least one node in the word network meets a preset condition and the calculation is stopped, and keywords with similar meanings are divided into corresponding topic clusters.
5. The user demand analysis method according to claim 2, characterized in that: in, The user data includes: user basic information and user vehicle information, and the data processing includes at least one of data cleaning, data integration, data transformation and data reduction.
6. The user demand analysis method according to claim 3, characterized in that: The determining of a plurality of keywords according to the TF-IDF value includes: Arrange all words in descending order according to the TF-IDF value; The words whose TF-IDF values are within the target range are selected as keywords.
7. A user demand analysis device, characterized in that: include: The acquisition module is used to obtain the user data and community chat data of private domain users of the target car company; An input module, used to input the community chat data into a target model, wherein the target model sequentially processes the community chat data using a TF-IDF algorithm and a clustering algorithm, and outputs the chat topic; An analysis module is used to analyze the needs of the private domain users based on the chat topic and the user data.
8. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the user demand analysis method according to any one of claims 1 to 6.
9. A computer-readable storage medium having a computer program or instruction stored thereon, characterized in that: The computer program or instruction is executed by a processor to implement the user demand analysis method as described in any one of claims 1-6.
10. A computer program product comprising a computer program or instructions, characterized in that When the computer program or instruction is executed, the user demand analysis method as described in any one of claims 1-6 is implemented.
Citation Information
Patent Citations
Chat type commodity shopping guide method and device, equipment and medium
CN115689672A
User focus analysis method and device, vehicle and storage medium
CN116050388A
Chat scene matching system, method and equipment based on context and storage medium
CN116881730A
User complaint mining method and device based on chat content, vehicle and medium
CN118260384A
KR20240007989A
