Combined recommendation scheme generation method and system based on product difference analysis

Abstract

The invention discloses a combined recommendation scheme generation method and system based on product difference analysis, and relates to the field of product comparison, and the method comprises the steps: obtaining the demand information of a customer, and carrying out the analysis to obtain a consultation product and a consultation item of a corresponding product; matching the consulting product with a stored product, and determining product information of the consulting product; the consulting product is judged, and the consulting condition is determined; if the consultation condition is single-product multi-parameter consultation, processing the product information of the consultation product to obtain display data; if the consultation condition is multi-product full-parameter consultation, performing data integration on the product information of the plurality of consultation products to obtain display data; if the consultation condition is multi-product multi-parameter consultation, extracting and summarizing product information of a plurality of consultation products to obtain display data; and performing data verification on the display data, and outputting the verified data. The method has the effect of improving the accuracy of product difference analysis.

Description

Technical Field

[0001] The present application relates to the field of product comparison, and in particular to a method and system for generating a combination recommendation solution based on product difference analysis. Background Art

[0002] With the rapid development of internet technology, e-commerce has become an integral part of people's daily lives. On e-commerce platforms, facing customer inquiries about multiple products, how to quickly compare the corresponding parameters of multiple products based on customer needs, so that customers can quickly compare the parameter differences between multiple products, has become an urgent problem to be solved.

[0003] In traditional technology, staff often retrieve corresponding product information based on customer needs, then filter various parameters in the product information based on the customer's needs, such as model, image, charging method, material, etc., and perform corresponding statistical comparisons on the filtered parameters to obtain a final comparison table of the corresponding parameter data of multiple products. The comparison table is then sent to the customer for review. However, when there are many customers or customer needs, the above solution requires a lot of time to filter and extract data and verify the organized data. This not only requires a lot of manpower and time, but is also prone to errors and needs improvement. Summary of the Invention

[0004] In order to improve the accuracy of product difference analysis, the present application provides a method and system for generating a combination recommendation solution based on product difference analysis.

[0005] In the first aspect, the present application provides a method for generating a combination recommendation solution based on product difference analysis, which adopts the following technical solution:

[0006] A method for generating a combination recommendation solution based on product difference analysis, comprising:

[0007] Obtain product information of each product from the user, perform attribute judgment on the product information of each product, and determine the attribute information of each parameter item in the product information;

[0008] Obtain customer demand information, make judgments on the customer demand information, clarify the products that the customer needs to consult and the corresponding product parameters, mark the products that need to be consulted as consulting products, and mark the corresponding product parameters that need to be consulted as consulting items;

[0009] Match the consulting product with the stored product to determine the product information of the consulting product;

[0010] Determine the consulting status based on the product category, number of inquiries, and number of consulting items for the corresponding consulting products; the consulting status includes single product multi-parameter consultation, single product full-parameter consultation, multi-product multi-parameter consultation, and multi-product full-parameter consultation;

[0011] If the inquiry is for a single product with full parameters, the product information of the inquiry product will be output;

[0012] If the inquiry is for a single product with multiple parameters, the product information of the inquiry product is processed according to the attribute information of the inquiry item and the parameter item to obtain the final display data;

[0013] If the inquiry is for multiple products with full parameters, the product information of multiple inquiries is integrated to obtain comparative display data;

[0014] If the inquiry status is a multi-product multi-parameter inquiry, then the product information of multiple inquiry products is extracted and summarized based on the attribute information of the inquiry items and parameter items, and the display data of the multiple inquiry products is obtained by comparing them with each other on the inquiry items;

[0015] Perform data verification on the displayed data and output the verified data.

[0016] Preferably, product information of each product of the user is obtained, and the correlation between the parameter items in each product information and the product is analyzed;

[0017] If the parameter item and the product are unique, it is determined that the parameter item and the product are associated, and the attribute information of the associated parameter item is marked as a display item;

[0018] The user's display setting information is obtained, and attribute information of non-related parameter items is marked as display items according to the display setting information; the display items are parameter items that must be displayed when the product information is displayed.

[0019] Preferably, the number of consultations for the consulting product is statistically determined. If the number of consultations for the consulting product is 1, the number of consultation items for the consulting product is statistically determined.

[0020] If the number of consulting items for the consulting product is 0, the customer's consulting status is determined to be a single product full parameter consultation;

[0021] If the number of consultation items for the consultation product is greater than 0, then the parameter items corresponding to the consultation items are matched with the parameter items of the corresponding product information to determine whether the parameter items corresponding to the consultation items are perfectly matched with the parameter items of the corresponding product information; a perfect match means that the number and types of the parameter items in the consultation items and the parameter items in the corresponding product information are the same;

[0022] If the parameter items corresponding to the consultation item perfectly match the parameter items of the corresponding product information, the customer's consultation status is determined to be a single product full-parameter consultation; if the parameter items corresponding to the consultation item do not perfectly match the parameter items of the corresponding product information, the customer's consultation status is determined to be a single product multi-parameter consultation;

[0023] If the number of consultations for a consulting product is greater than 1, the product category of the consulting product is determined based on the product information of the consulting product, and the consulting products are divided according to the category, and the divided consulting products are marked as consulting products of the same category;

[0024] Conduct statistical judgment on the number of consultations for consulting products of the same category. If the number of consultations for consulting products of the same category is greater than 1, then judge the number of consultation items for each consulting product in the same category;

[0025] If the number of inquiries for the same type of consulting product is greater than 1, and the number of consulting items for each consulting product in the same type of consulting product is 0, or the parameter items corresponding to the consulting items perfectly match the parameter items of the corresponding product information, then the customer's consulting status is determined to be a multi-product full-parameter consultation;

[0026] If the number of consultations for consulting products of the same category is greater than 1, and the number of consultation items for each consulting product in the same category is not perfectly matched, the customer's consulting status is determined to be multi-product multi-parameter consultation.

[0027] Preferably, the product information of the consulting product is copied to obtain information to be processed;

[0028] Matching the consultation item with the information to be processed to determine whether the parameter item in the consultation item completely matches the information to be processed;

[0029] If all the parameters in the inquiry item can match the parameters in the information to be processed, the successfully matched parameters in the information to be processed will be marked;

[0030] Based on the attribute information of the parameter items in the information to be processed, marking the parameter items whose attribute information is a display item;

[0031] Delete the unmarked parameter items and their data in the information to be processed to obtain the final display data.

[0032] Preferably, any one of the product information corresponding to the multiple consulting products is selected, and the selected product information is used as the initial display data;

[0033] Matching the parameter items of the product information of the remaining products with the parameter items of the initial display data, marking the successfully matched parameter items in the product information of the remaining products to obtain first marked data, and marking the position of the successfully matched parameter items in the initial display data to obtain a marked position;

[0034] Record the first mark data accordingly according to the mark position, and sort them out to obtain intermediate display data;

[0035] Mark the unmatched parameter items in the product information of the remaining products to obtain a first marking parameter and its corresponding second marking data;

[0036] The first marking parameter is added to the blank data of the intermediate display data to obtain the position information of the first marking parameter, and the second marking data is entered according to the position information of the first marking parameter to obtain the final display data.

[0037] Preferably, a blank display file is created based on the built-in display template; the blank display file includes a project column and a data column;

[0038] Aggregate the consulting items of multiple consulting products to obtain a total set of consulting items;

[0039] Filling the item column in the blank display file based on the total set of consultation items to obtain a first display file;

[0040] Match the consulting items of each consulting product with the parameter items in the product information corresponding to the consulting product, mark the successfully matched parameter items in the product information to obtain a second marking parameter, mark the corresponding parameter data as third marking data, mark the consulting items in the product information that failed to match to obtain a third marking parameter, and record the data corresponding to the third marking parameter as empty data;

[0041] Matching the consulting items of each consulting product with the item column in the first display file to obtain the location information of the corresponding consulting item;

[0042] The second marking parameter and the third marking parameter of each consulting product are recorded into the data column of the first display file according to the position information of the corresponding consulting item, thereby obtaining a second display file;

[0043] The attribute information of the parameter items in the product information of each consulting product is judged, and the parameter items whose attribute information is the display item are matched with the project column in the second display file. If the project column in the second display file and the parameter items whose attribute information is the display item are not perfectly matched, the unmatched parameter items whose attribute information is the display item are added to the second display file to obtain the final display data.

[0044] Preferably, the displayed data is read and judged to determine the product corresponding to the displayed data, and marked as a recorded product;

[0045] Match the recorded product with the consulted product to determine whether the recorded product is the same as the consulted product. If the recorded product is different from the consulted product, the displayed data is determined to be inaccurate and the displayed data is regenerated.

[0046] If it is determined that the recorded product is the same as the queried product, then the product information corresponding to the recorded product is determined based on the recorded product and the displayed data;

[0047] Matching the product information of the recorded product with the product information of the corresponding consulting product to determine whether the product information of the recorded product completely matches the product information of the consulting product; if the product information of the recorded product does not completely match the product information of the consulting product, determining that the displayed data is inaccurate and regenerating the displayed data;

[0048] If the product information of the recorded product completely matches the product information of the consulted product, then the parameter items that successfully match are recorded according to the product information of the recorded product to obtain the parameter item sequence;

[0049] Comparing the order of parameter items of multiple recorded products to determine whether the order of parameter items of multiple recorded products is the same; if the order of parameter items of multiple recorded products is different, determining that the displayed data is inaccurate, and regenerating the displayed data;

[0050] If the order of the parameter items of the multiple recorded products is the same, it is determined that the displayed data is accurate and the displayed data is output.

[0051] In a second aspect, the present application provides a system for generating combination recommendation solutions based on product difference analysis, which adopts the following technical solutions:

[0052] A combination recommendation solution generation system based on product difference analysis includes: a data storage module, a demand judgment module, a data analysis module and a data verification module;

[0053] The data storage module obtains the product information of each product of the user, and performs attribute judgment on the product information of each product to determine the attribute information of each parameter item in the product information;

[0054] The demand judgment module obtains and judges the customer's demand information, clarifies the products that the customer needs to consult and the corresponding product parameters, and marks the products that need to be consulted as consulting products and the corresponding product parameters that need to be consulted as consulting items;

[0055] The data analysis module matches the consulting product with the stored products to determine the product information of the consulting product; judges the variety category of the consulting product, the number of consultations and the number of consultation items of the corresponding consulting product to determine the consulting status; the consulting status includes single-product multi-parameter consultation, single-product full-parameter consultation, multi-product multi-parameter consultation and multi-product full-parameter consultation; if the consulting status is single-product full-parameter consultation, the product information of the consulting product is output; if the consulting status is single-product multi-parameter consultation, the product information of the consulting product is processed according to the attribute information of the consulting items and parameter items to obtain the final display data; if the consulting status is multi-product full-parameter consultation, the product information of multiple consulting products is integrated to obtain display data with mutual comparison; if the consulting status is multi-product multi-parameter consultation, the product information of multiple consulting products is extracted and summarized according to the attribute information of the consulting items and parameter items to obtain display data of multiple consulting products that are compared with each other on the consulting items;

[0056] The data verification module verifies the displayed data and outputs the verified data.

[0057] In summary, this application includes at least one of the following beneficial technical effects:

[0058] 1. Obtain the product information of each product from the user to determine the data in the database. By judging the customer's needs, the customer's needs are clarified, which helps in the subsequent integration of product differences and makes the integration results effective. Then, by matching the customer's needs with the product information input by the user into the system, and re-judging the matched product information and the customer's consulting products, the relationship between the products that the customer needs to consult and the needs of the consulting products themselves are further clarified, so as to determine whether the product information of the consulting products needs to be reprocessed to accurately provide product data for customers to view. Different technical means are used according to different consulting situations to improve the accuracy and convenience of data integration. The integrated data is further verified to ensure the accuracy of the final output results, thereby improving the accuracy of product difference analysis.

[0059] 2. By selecting any one of the product information of multiple inquiries, the parameter item data corresponding to the initial display data is accurate. Then, by matching the parameter items of the product information of the remaining products with the parameter items of the initial display data, the storage location of the data of each parameter item of the remaining products is clarified, and the data of the corresponding location is recorded, ensuring the correctness and accuracy of data storage. At the same time, when the parameter items contained in the product information of the remaining products exceed the parameter items in the intermediate display data, the parameter items are added to the intermediate display data to ensure the integrity of the product information of each inquiries product. The final display data can effectively display the product information of the product inquired by the customer, improving the accuracy of product difference analysis;

[0060] 3. By reading the integrated display data, it is determined whether there is any data omission in the data integration process, thereby ensuring the integrity of the data. By comparing the information of a single recorded product with the product information of the corresponding consulting product, it is determined that the recorded data all comes from the corresponding consulting product, thereby ensuring the correctness of data entry. Finally, by comparing the matching order between the information of multiple recorded products and the product information of the corresponding consulting products, it is determined that the recording position of the data of each recorded product is correct. Then, through three verifications, the accuracy of the data in the final display data is guaranteed, thereby improving the accuracy of product difference analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 This is a flowchart of the steps of the method for generating a combination recommendation solution based on product difference analysis in this embodiment;

[0062] Figure 2 This is a module block diagram of the system for generating combination recommendation solutions based on product difference analysis in this embodiment. Figure numerals: 1. Data storage module; 2. Demand judgment module; 3. Data analysis module; 4. Data verification module. DETAILED DESCRIPTION

[0063] The following is combined with Figure 1-Figure 2 This application is described in further detail.

[0064] The embodiments of the present application disclose a method and system for generating a combination recommendation solution based on product difference analysis.

[0065] Example: Figure 1 As shown, the present invention provides a method for generating a combination recommendation solution based on product difference analysis, comprising:

[0066] S100, obtaining product information of each product of the user, and performing attribute judgment on the product information of each product to determine the attribute information of each parameter item in the product information;

[0067] S200, obtaining customer demand information, determining the customer demand information, identifying the product the customer needs to consult and the corresponding product parameters, marking the product to be consulted as a consulting product, and marking the corresponding product parameters to be consulted as consulting items;

[0068] S300, matching the consulting product with the stored products to determine product information of the consulting product;

[0069] S400: Determine the product category, number of inquiries, and number of inquiries for the corresponding product to determine the inquiry status; the inquiry status includes single product multi-parameter inquiry, single product full-parameter inquiry, multi-product multi-parameter inquiry, and multi-product full-parameter inquiry.

[0070] S500: If the inquiry status is for a single product full parameter inquiry, the product information of the inquiry product is output;

[0071] S600, if the inquiry status is a single product multi-parameter inquiry, the product information of the inquiry product is processed according to the attribute information of the inquiry item and the parameter item to obtain the final display data;

[0072] S700, if the inquiry status is a multi-product full parameter inquiry, then data integration is performed on the product information of the multiple inquiry products to obtain display data with mutual comparison;

[0073] S800, if the inquiry status is a multi-product multi-parameter inquiry, extract and aggregate product information of multiple inquiry products based on the attribute information of the inquiry items and parameter items to obtain display data comparing the multiple inquiry products with respect to the inquiry items;

[0074] S900: Verify the displayed data and output the verified data.

[0075] In this embodiment, the data in the database is determined by obtaining the product information of each product of the user, and the customer's needs are clarified by judging the customer's needs, which provides assistance for the subsequent product difference integration and makes the integration result valid. The customer's needs are then matched with the product information input by the user into the system, and the matched product information and the customer's consulting products are re-judged to further clarify the relationship between the products that the customer needs to consult, as well as the needs for the consulting products themselves, so as to clarify whether the product information of the consulting products needs to be reprocessed to accurately display the product data for customers to view. Different technical means are used according to different consulting situations to improve the accuracy and convenience of data integration, and the integrated data is verified to further ensure the accuracy of the final output results, thereby improving the accuracy of product difference analysis.

[0076] For example, the customer adds the product he wants to inquire about and the parameter items of the corresponding product through the operation interface. For example, the product information of product A includes parameter items a, b, c, and d, and the product information of product B includes a, b, c, and d. First, the number of products the customer inquires about is judged. If the customer only inquires about product A, there is no need to integrate the product information with each other, so it is only necessary to judge whether the inquired parameter items are all the information of the product. If it is all the information, it can be output directly. Otherwise, if not, the information needs to be processed. Similarly, when the customer inquires about product A and product B at the same time, it is necessary to determine whether product A and product B are the same type of products. For example, product A is a mobile phone and product B is an aromatherapy, it indicates that the two products are of different categories. On the contrary, when products A and B are both mobile phones, it indicates that the two products are the same product. For non-similar products, there is no need to compare the parameter item data of A and B with each other. On the contrary, for similar products, it is necessary to compare the parameter item data of A and the parameter item data of B, and then integrate them into the final display data. The display data is then verified to determine the correctness of the integration process, thereby improving the accuracy of product difference analysis.

[0077] In step S100, the product information of each product of the user is obtained, and the attribute of the product information of each product is judged to determine the attribute information of each parameter item in the product information, including the following steps:

[0078] S110, obtaining product information of each product of the user, and comparing the correlation between the parameter items in each product information and the product;

[0079] S120, if the parameter item is unique to the product, it is determined that the parameter item is associated with the product, and the attribute information of the associated parameter item is marked as a display item; wherein the uniqueness is a unique identification of the product category, such as the product model, picture, name, etc.

[0080] S130: Obtain the user's display setting information and, based on the display setting information, mark the attribute information of the non-related parameter items as display items; the display items are parameter items that must be displayed when displaying product information. Manual settings can be used to determine the information to be displayed to effectively identify the information and reduce recognition errors.

[0081] In this embodiment, by analyzing the parameter items in the product information, determining the uniqueness between the parameter items and the product, and obtaining the display setting information manually set by the user, the parameter items that must be displayed in the product information are clarified, so that accurate and effective product information can be provided to customers when they consult about product information, thereby improving the effective recognition of the displayed data and reducing the customer's recognition barriers to product information.

[0082] For example, by setting display items to assist customers in effectively identifying product information, for example, when viewing product A and product B simultaneously, the output information is tabular data, with one column containing various data items for product A and another for product B. However, because unique information such as product name, model, and image is not indicated, customers cannot clearly know which column contains various data items for product A and which column contains various data items for product B. This creates obstacles for customer identification and reduces the customer experience.

[0083] In step S400, the type of consulting product, the number of consultations, and the number of consultation items corresponding to the consulting product are judged to determine the consulting status, including the following steps:

[0084] S410, performing statistical determination on the number of consultations for the consulting product. If the number of consultations for the consulting product is 1, performing statistical determination on the number of consultation items for the consulting product.

[0085] S420, if the number of consultation items for the consultation product is 0, it is determined that the customer's consultation status is a single product full parameter consultation;

[0086] S430: If the number of inquiry items for the inquiry product is greater than 0, then matching the parameter items corresponding to the inquiry items with the parameter items of the corresponding product information to determine whether the parameter items corresponding to the inquiry items perfectly match the parameter items of the corresponding product information; a perfect match means that the number and types of the parameter items in the inquiry items and the parameter items in the corresponding product information are the same;

[0087] S440: If the parameter items corresponding to the inquiry item perfectly match the parameter items of the corresponding product information, the customer's inquiry status is determined to be a single product full-parameter inquiry; if the parameter items corresponding to the inquiry item do not perfectly match the parameter items of the corresponding product information, the customer's inquiry status is determined to be a single product multi-parameter inquiry;

[0088] S450: If the number of consultations for the consulting product is greater than 1, determine the product category of the consulting product based on the product information of the consulting product, classify the consulting products according to the category, and mark the classified consulting products as consulting products of the same category;

[0089] S460, statistically determine the number of consultations for consulting products of the same category. If the number of consultations for consulting products of the same category is greater than 1, determine the number of consultation items for each consulting product in the same category.

[0090] S470: If the number of inquiries for the same type of consulting product is greater than 1, and the number of consulting items for each consulting product in the same type of consulting product is 0, or the parameter item corresponding to the consulting item perfectly matches the parameter item of the corresponding product information, then the customer's consulting status is determined to be a multi-product full-parameter consultation;

[0091] S480: If the number of consultations for consulting products of the same category is greater than 1, and the number of consultation items for each consulting product in the same category does not perfectly match, then the customer's consulting status is determined to be multi-product multi-parameter consultation.

[0092] In this embodiment, the consulting data of the customer's consulting product is first judged to determine whether it is necessary to judge the variety category of the consulting product, and then clarify the category to which the consulting situation belongs. Then, by judging the number of consulting items of each product and the matching between the consulting items and the product information, it is determined whether the customer's consulting situation on the product is consulting on all parameter items of the product. Then, based on the number of products consulted in the same category and the consulting situation on each product parameter item, the corresponding consulting situation category is clarified. Different consulting situations have different requirements for display data, so corresponding processing measures can be adopted according to different consulting situations, so that the final display data is more accurate and effective.

[0093] For example, when a customer only inquires about product A, it indicates that there is no need to integrate the product information of multiple products, and only the required data extraction of the product information of product A is required. If the customer does not indicate that he wants to view a specific parameter item of product A, for example: the customer's demand is to view the details of product A, it indicates that all parameter items of product A are to be viewed. Similarly, when the customer clearly indicates that he wants to view certain parameter items of product A, the parameter items corresponding to the product information are matched to determine whether the customer's demand is the same as the product information. If they are the same, it indicates that all parameter items of product A are to be viewed, otherwise not. Similarly, when the customer views information on products A and B at the same time, the variety category of products A and B is determined to determine whether they are indeed divided into two separate individuals, and then the above judgment is made on the consultation items of products A and B.

[0094] In step S600, if the inquiry status is a single product multi-parameter inquiry, the product information of the inquiry product is processed according to the attribute information of the inquiry items and parameter items to obtain the final display data, including the following steps:

[0095] S610, copying the product information of the consulting product to obtain information to be processed;

[0096] S620, matching the consultation item with the information to be processed, and determining whether the parameter item in the consultation item completely matches the information to be processed;

[0097] S630, if all the parameter items in the inquiry item match the parameter items in the information to be processed, then the successfully matched parameter items in the information to be processed are marked;

[0098] S640, marking the parameter items whose attribute information is a display item based on the attribute information of the parameter items in the information to be processed;

[0099] S650: Delete the unmarked parameter items and their data in the information to be processed to obtain final display data.

[0100] In this embodiment, the security and accuracy of the original data are guaranteed by copying the product information corresponding to the consulting product, and the parameter items that are successfully matched in the product information are marked, as well as the parameter items whose attribute information is a display item, so as to clarify the data to be retained in the final display data. Then, the unmarked parameter items and their data in the information to be processed are deleted to avoid data changes to the required data, thereby ensuring the integrity and accuracy of the final display data.

[0101] For example, when it is determined that the consultation status is a single product multi-parameter consultation, it indicates that effective data extraction of the product information of the product is required. For example, the consultation items of consultation product A are a and b, and the product information corresponding to consultation product A includes a, b, c, and d. Then, the data of items a and b in the product information need to be extracted to reduce the interference of unnecessary information c and d. By matching the consultation items a and b with the product information a, b, c, and d, it is determined that a and b in the product information are required, and then items a and b in the product information are marked. Then, the attribute information of the remaining unnecessary information items c and d is judged to determine whether there are display items in items c and d. If the attribute information of c is a display item, it indicates that item d is a redundant item. Then, by deleting item d and its corresponding data value in the copied information to be processed, the consultation item required by the customer is obtained. By deleting unnecessary data, changes in required requirements are avoided, and the final displayed data is consistent with the original product information, thereby improving the accuracy of the data.

[0102] In step S700, if the consultation status is a multi-product full parameter consultation, the product information of the multiple consultation products is integrated to obtain display data with mutual comparison, including the following steps:

[0103] S710, selecting any one of the product information corresponding to the multiple consulting products, and using the selected product information as initial display data;

[0104] S720, matching the parameter items of the product information of the remaining products with the parameter items of the initial display data, marking the successfully matched parameter items in the product information of the remaining products to obtain first marked data, and marking the position of the successfully matched parameter items in the initial display data to obtain a marked position;

[0105] S730, recording the first mark data according to the mark position, and sorting them to obtain intermediate display data;

[0106] S740: Mark the unmatched parameter items in the product information of the remaining products to obtain a first marking parameter and its corresponding second marking data;

[0107] S750: Add the first marking parameter to the blank data of the intermediate display data to obtain the position information of the first marking parameter, and enter the second marking data according to the position information of the first marking parameter to obtain the final display data.

[0108] In this embodiment, by selecting any one of the product information of multiple consulting products, the parameter item data corresponding to the initial display data is accurate, and then by matching the parameter items of the product information of the remaining products with the parameter items of the initial display data, the location where the data of each parameter item in the remaining products needs to be stored is clarified, and the data at the corresponding location is recorded, thereby ensuring the correctness and accuracy of data storage. At the same time, when the parameter items contained in the product information of the remaining products exceed the parameter items in the intermediate display data, the integrity of the product information of each consulting product is guaranteed by adding parameter items to the intermediate display data, so that the final display data can effectively display the product information of the product consulted by the customer, thereby improving the accuracy of product difference analysis.

[0109] For example, when the consultation status is determined to be a multi-product full-parameter consultation, it indicates that the product information of multiple products needs to be regularized to integrate into a complete display data, without the need to extract the product information of a single product. For example, the consultation items for consultation product A are a, b, c, d, and the consultation items for consultation product B are a, b, c, d. If the product information of consultation products A and B are both a, b, c, d, then consultation product A or consultation product B is randomly selected. Assuming that product A is selected, the product information of product A is used as the initial display data. For example, assuming that the data is stored in a table, the distribution form of the initial display data is that the first row and first column is parameter item a, the second row and first column is parameter item b, and so on. The first row and second column are the data values ​​of parameter item a of product A, the second row and second column are the data values ​​of parameter item b of product A, and so on. The second column is marked as the storage location. For product A, the data storage location of parameter item a is (2,1), the storage location of parameter item b is (2,2), and so on. By matching the product information a, b, c, and d of consulting product B with the initial display data, the storage locations of the corresponding parameter items are determined. The matching result for consulting product B is that the data storage location of parameter item a is (3, 1), the storage location of parameter item b is (3, 2), and so on. If the consulting items of consulting product B are a, b, c, d, and e, and the product information of consulting product B is a, b, c, d, and e, the storage locations of items a, b, c, and d are determined through matching, but the storage location of item e is not matched. Parameter item e is then added to the initial display data, resulting in intermediate display data containing the five parameter items a, b, c, d, and e. The unmatched item e of consulting product B is then rematched, and the storage location of parameter item e is determined to be (3, 5). The result is the final display data with the first column containing parameter items a, b, c, d, and e, the second column containing the data values ​​of product A for the corresponding parameter items, and the third column containing the data values ​​of product B for the corresponding parameter items.

[0110] In step S800, if the inquiry status is a multi-product multi-parameter inquiry, product information of multiple inquiry products is extracted and aggregated based on the attribute information of the inquiry items and parameter items to obtain display data of the multiple inquiry products compared with each other on the inquiry items, including the following steps:

[0111] S810, creating a blank display file based on the built-in display template; the blank display file includes a project column and a data column;

[0112] S820, summarizing the consulting items of multiple consulting products to obtain a total set of consulting items;

[0113] S830, filling the item column in the blank display file based on the total set of consultation items to obtain a first display file;

[0114] S840: Match the consulting items of each consulting product with the parameter items in the product information corresponding to the consulting product, mark the successfully matched parameter items in the product information to obtain a second marking parameter, mark the corresponding parameter data as third marking data, mark the consulting items in the product information that failed to match to obtain a third marking parameter, and record the data corresponding to the third marking parameter as empty data;

[0115] S850, matching the consulting items of each consulting product with the item column in the first display file to obtain location information of the corresponding consulting item;

[0116] S860, recording the second marking parameter and the third marking parameter of each consulting product according to the position information of the corresponding consulting item, and the third marking data and the empty data into the data column of the first display file to obtain a second display file;

[0117] S870, judge the attribute information of the parameter items in the product information of each consulting product, match the parameter items whose attribute information is the display item with the project column in the second display file, if the project column in the second display file and the parameter items whose attribute information is the display item are not perfectly matched, then add the unmatched parameter items whose attribute information is the display item to the second display file to obtain the final display data.

[0118] In this embodiment, a blank display file is created by using a built-in display template, and the created blank display file is divided to clarify the storage location of the corresponding consultation item and the storage location of the corresponding consultation item data value, making data storage more organized. At the same time, the created blank display file is filled with consultation items to further clarify the storage location corresponding to the data value of each consultation item, which facilitates data management. By matching the consultation items of the consultation product with the product information of the consultation product, and storing the data values ​​of the successfully matched parameter items in corresponding locations, the product information of the consultation product can be correctly stored. At the same time, blank data is filled for the consultation items that are not successfully matched in the database, thereby reducing data storage errors for the same consultation item for multiple consultation products, thereby making the final display data more accurate and effective. For example, if parameter item g is consulted for both products A and B, but product A does not have parameter item g, and product B has parameter item g, then if parameter item g of product A is not filled with data, the data value of parameter item g of product B may be stored in the storage location of parameter item g of product A, resulting in data misalignment, thereby affecting the analysis of product differences.

[0119] For example, when it is determined that the consultation status is a multi-product multi-parameter consultation, it means that not only the product information of the products consulted by the customer needs to be integrated, but also the product information of the consulted products needs to be effectively extracted to ensure that the final displayed data is accurate and valid. By establishing a blank display file and statistically entering the parameter items that need to be consulted, a first display file with parameter item information is obtained. For example, a blank table is established according to the built-in display template. The sum of the consultation items of multiple consultation products is a, b, c, d, e, f, g. Then a, b, c, d, e, f, g are input into the blank table to establish a first display file with consultation item information. Then, the first display file with consultation item information is filled with data of corresponding parameter items. By matching the consultation items with the product information, the data value of the corresponding consultation item is clarified. For example, the consultation items for product A are parameter items a and b, and it is determined that product A is obtained. The data values ​​of parameter items a and b are Aa and Ab, respectively. By matching parameter items a and b in the first display file, for example, if the location information of parameter item a is (2,1), the data value Aa of parameter item a is added to the storage location (2,1). If the consultation item also includes g, but the product information of product A does not include parameter item g, the data value of parameter item g is marked as empty data. For example, if the location information of parameter item g is (2,7), the data value of parameter item g: empty data is added to the storage location (2,7). After the consultation items of each consultation product are added to the first display file, the added first display file is marked as the second display file. At the same time, the attribute information of the parameter items in the product information of each consultation product is judged, and the data of the parameter items not added to the second display file is added to ensure data integrity.

[0120] In step S900, the display data is verified and the verified data is output, which includes the following steps:

[0121] S910, performing data reading and judgment on the displayed data, determining the product corresponding to the displayed data, and marking it as a recorded product;

[0122] S920, matching the recorded product with the consulted product to determine whether the recorded product is the same as the consulted product. If the recorded product is different from the consulted product, the displayed data is determined to be inaccurate and the displayed data is regenerated.

[0123] S930, if it is determined that the recorded product is the same as the consulted product, determining product information corresponding to the recorded product based on the recorded product and the displayed data;

[0124] S940, matching the product information of the recorded product with the product information of the corresponding consulting product to determine whether the product information of the recorded product completely matches the product information of the consulting product. If the product information of the recorded product does not completely match the product information of the consulting product, determining that the displayed data is inaccurate and regenerating the displayed data;

[0125] S950, if the product information of the recorded product completely matches the product information of the consulting product, then the successfully matched parameter items are recorded according to the product information of the recorded product to obtain the parameter item sequence; wherein, a complete match means that the product information in the recorded product can all be found in the product information of the consulting product.

[0126] S960, comparing the order of the parameter items of the multiple recorded products to determine whether the order of the parameter items of the multiple recorded products is the same; if the order of the parameter items of the multiple recorded products is different, determining that the display data is inaccurate, and regenerating the display data;

[0127] S970: If the order of the parameter items of the multiple recorded products is the same, it is determined that the displayed data is accurate, and the displayed data is output.

[0128] In this embodiment, data is read from the integrated display data to determine whether there is any data omission during the data integration process, thereby ensuring the integrity of the data. The information of a single recorded product is compared with the product information of the corresponding consulting product to determine that the recorded data all comes from the corresponding consulting product, thereby ensuring the correctness of data entry. Finally, the matching order between the information of multiple recorded products and the product information of the corresponding consulting product is compared to determine that the recording position of the data of each recorded product is correct. Then, through three checks, the accuracy of the data in the final display data is guaranteed, thereby improving the accuracy of product difference analysis.

[0129] For example, the displayed data records products A, B, and C. If the customer also inquires about products A, B, and C, it means that the products the customer wants to inquire about have been recorded in the displayed data. Therefore, the next step is to judge the correctness of the data record. The correctness judgment of the data record includes the correctness of the data value and the correctness of the data record position. By matching the product information of the same product in the displayed data with the product information in the database, it is determined whether the product information in the displayed data all comes from the product information in the database, and then the correctness of the data record is judged. After determining the correctness of the data record, since the record order of each parameter item of each product in the integrated display data is the same, by making the product information of each product in the displayed data and the product information in the database match in the same order, it shows that the data record order of each product is the same, thereby proving the correctness of the data record position. For example, the theoretical recording order should be parameter items a, b, c, d. The data values ​​for product A are Aa, Ab, Ac, Ad, and the data values ​​for product B are Ba, Bb, Bc, Bd. However, if in actual recording, the data value recording order of product B is Ba, Bc, Bb, Bd, then the parameter item order between product A and product B will be different, and it is determined that there is an error in the data recording position of product A and / or product B, and thus it is determined that there is an error in the displayed data.

[0130] Based on the description of the embodiment of the method for generating a combination recommendation solution based on product difference analysis, the embodiment of the present invention further discloses a system for generating a combination recommendation solution based on product difference analysis:

[0131] like Figure 2 As shown, a system for generating a combination recommendation solution based on product difference analysis, by applying the above-mentioned method for generating a combination recommendation solution based on product difference analysis, comprises: a data storage module 1, a demand judgment module 2, a data analysis module 3 and a data verification module 4;

[0132] Data storage module 1, obtains product information of each product of the user, and performs attribute judgment on the product information of each product to determine the attribute information of each parameter item in the product information;

[0133] Demand judgment module 2 obtains customer demand information, judges the customer demand information, clarifies the products that the customer needs to consult and the corresponding product parameters, and marks the products that need to be consulted as consulting products and the corresponding product parameters that need to be consulted as consulting items;

[0134] Data analysis module 3 matches the consulting product with the stored products to determine the product information of the consulting product; judges the variety category of the consulting product, the number of consultations and the number of consultation items of the corresponding consulting product to determine the consulting status; the consulting status includes single product multi-parameter consultation, single product full-parameter consultation, multi-product multi-parameter consultation and multi-product full-parameter consultation; if the consulting status is single product full-parameter consultation, the product information of the consulting product is output; if the consulting status is single product multi-parameter consultation, the product information of the consulting product is processed according to the attribute information of the consulting items and parameter items to obtain the final display data; if the consulting status is multi-product full-parameter consultation, the product information of multiple consulting products is integrated to obtain display data with mutual comparison; if the consulting status is multi-product multi-parameter consultation, the product information of multiple consulting products is extracted and summarized according to the attribute information of the consulting items and parameter items to obtain display data of multiple consulting products that are compared with each other on the consulting items;

[0135] The data verification module 4 verifies the display data and outputs the verified data.

[0136] Compared with the existing method and system for generating combination recommendation solutions based on product difference analysis, the present invention improves the accuracy of product difference analysis.

[0137] The above are all preferred embodiments of the present application, and are not intended to limit the scope of protection of the present application. Therefore, any equivalent changes made based on the structure, shape, and principle of the present application should be included in the scope of protection of the present application.

Claims

1. A method for generating a combination recommendation solution based on product difference analysis, characterized in that: include: Obtain product information of each product from the user, perform attribute judgment on the product information of each product, and determine the attribute information of each parameter item in the product information; Obtain customer demand information, make judgments on the customer demand information, clarify the products that the customer needs to consult and the corresponding product parameters, mark the products that need to be consulted as consulting products, and mark the corresponding product parameters that need to be consulted as consulting items; Match the consulting product with the stored product to determine the product information of the consulting product; Determine the type and quantity of consulting products, the number of consulting items and the number of consulting items for the corresponding consulting products, and determine the consulting status; The consultation conditions include single product multi-parameter consultation, single product full parameter consultation, multi-product multi-parameter consultation and multi-product full parameter consultation; If the inquiry is for a single product with full parameters, the product information of the inquiry product will be output; If the inquiry is for a single product with multiple parameters, the product information of the inquiry product is processed according to the attribute information of the inquiry item and the parameter item to obtain the final display data; If the inquiry is for multiple products with full parameters, the product information of multiple inquiries is integrated to obtain comparative display data; If the inquiry status is a multi-product multi-parameter inquiry, then the product information of multiple inquiry products is extracted and summarized based on the attribute information of the inquiry items and parameter items, and the display data of the multiple inquiry products is obtained by comparing them with each other on the inquiry items; Perform data verification on the displayed data and output the verified data.

2. The method for generating a combination recommendation solution based on product difference analysis according to claim 1, characterized in that: The method of obtaining product information of each product of the user and performing attribute judgment on the product information of each product to determine the attribute information of each parameter item in the product information is as follows: Obtain product information of each product of the user, and compare the correlation between the parameter items in each product information and the product; If the parameter item and the product are unique, it is determined that the parameter item and the product are associated, and the attribute information of the associated parameter item is marked as a display item; The user's display setting information is obtained, and attribute information of non-related parameter items is marked as display items according to the display setting information; the display items are parameter items that must be displayed when the product information is displayed.

3. The method for generating a combination recommendation solution based on product difference analysis according to claim 1, characterized in that: The consulting status is determined by judging the type of consulting product, the number of consultations, and the number of consultation items for the corresponding consulting product, specifically: Perform statistical judgment on the number of consultations for the consulting product. If the number of consultations for the consulting product is 1, perform statistical judgment on the number of consultation items for the consulting product. If the number of consulting items for the consulting product is 0, the customer's consulting status is determined to be a single product full parameter consultation; If the number of consultation items for the consultation product is greater than 0, then the parameter items corresponding to the consultation items are matched with the parameter items of the corresponding product information to determine whether the parameter items corresponding to the consultation items are perfectly matched with the parameter items of the corresponding product information; a perfect match means that the number and types of the parameter items in the consultation items and the parameter items in the corresponding product information are the same; If the parameter items corresponding to the consultation item perfectly match the parameter items of the corresponding product information, the customer's consultation status is determined to be a single product full parameter consultation; If the parameter items corresponding to the inquiry item do not perfectly match the parameter items of the corresponding product information, the customer's inquiry status is determined to be a single product multi-parameter inquiry; If the number of consultations for a consulting product is greater than 1, the product category of the consulting product is determined based on the product information of the consulting product, and the consulting products are divided according to the category, and the divided consulting products are marked as consulting products of the same category; Conduct statistical judgment on the number of consultations for consulting products of the same category. If the number of consultations for consulting products of the same category is greater than 1, then judge the number of consultation items for each consulting product in the same category; If the number of inquiries for the same type of consulting product is greater than 1, and the number of consulting items for each consulting product in the same type of consulting product is 0, or the parameter items corresponding to the consulting items perfectly match the parameter items of the corresponding product information, then the customer's consulting status is determined to be a multi-product full-parameter consultation; If the number of consultations for consulting products of the same category is greater than 1, and the number of consultation items for each consulting product in the same category is not perfectly matched, the customer's consulting status is determined to be multi-product multi-parameter consultation.

4. The method for generating a combination recommendation solution based on product difference analysis according to claim 1, characterized in that: If the consultation status is a single product multi-parameter consultation, the product information of the consultation product is processed according to the attribute information of the consultation item and the parameter item to obtain the final display data, specifically: Copy the product information of the consulting product to obtain the information to be processed; Matching the consultation item with the information to be processed to determine whether the parameter item in the consultation item completely matches the information to be processed; If all the parameters in the inquiry item can match the parameters in the information to be processed, the successfully matched parameters in the information to be processed will be marked; Based on the attribute information of the parameter items in the information to be processed, marking the parameter items whose attribute information is a display item; Delete the unmarked parameter items and their data in the information to be processed to obtain the final display data.

5. The method for generating a combination recommendation solution based on product difference analysis according to claim 4, characterized in that: If the consultation status is a multi-product full parameter consultation, the product information of the multiple consultation products is integrated to obtain display data with mutual comparison, specifically: Select any one of the product information corresponding to the multiple consulting products, and use the selected product information as the initial display data; Matching the parameter items of the product information of the remaining products with the parameter items of the initial display data, marking the successfully matched parameter items in the product information of the remaining products to obtain first marked data, and marking the position of the successfully matched parameter items in the initial display data to obtain a marked position; Record the first mark data accordingly according to the mark position, and sort them out to obtain intermediate display data; Mark the unmatched parameter items in the product information of the remaining products to obtain a first marking parameter and its corresponding second marking data; The first marking parameter is added to the blank data of the intermediate display data to obtain the position information of the first marking parameter, and the second marking data is entered according to the position information of the first marking parameter to obtain the final display data.

6. The method for generating a combination recommendation solution based on product difference analysis according to claim 5, characterized in that: If the consultation status is a multi-product multi-parameter consultation, then the product information of multiple consultation products is extracted and summarized based on the attribute information of the consultation items and parameter items to obtain display data of the multiple consultation products compared with each other on the consultation items, specifically: Based on the built-in display template, a blank display file is created; the blank display file includes a project column and a data column; Aggregate the consulting items of multiple consulting products to obtain a total set of consulting items; Filling the item column in the blank display file based on the total set of consultation items to obtain a first display file; Match the consulting items of each consulting product with the parameter items in the product information corresponding to the consulting product, mark the successfully matched parameter items in the product information to obtain a second marking parameter, mark the corresponding parameter data as third marking data, mark the consulting items in the product information that failed to match to obtain a third marking parameter, and record the data corresponding to the third marking parameter as empty data; Matching the consulting items of each consulting product with the item column in the first display file to obtain the location information of the corresponding consulting item; The second marking parameter and the third marking parameter of each consulting product are recorded into the data column of the first display file according to the position information of the corresponding consulting item, thereby obtaining a second display file; The attribute information of the parameter items in the product information of each consulting product is judged, and the parameter items whose attribute information is the display item are matched with the project column in the second display file. If the project column in the second display file and the parameter items whose attribute information is the display item are not perfectly matched, the unmatched parameter items whose attribute information is the display item are added to the second display file to obtain the final display data.

7. The method for generating a combination recommendation solution based on product difference analysis according to claim 6, characterized in that: The data verification of the display data and the output of the verified data are specifically as follows: Read and judge the displayed data, determine the product corresponding to the displayed data, and mark it as a recorded product; Match the recorded product with the consulted product to determine whether the recorded product is the same as the consulted product. If the recorded product is different from the consulted product, the displayed data is determined to be inaccurate and the displayed data is regenerated. If it is determined that the recorded product is the same as the queried product, then the product information corresponding to the recorded product is determined based on the recorded product and the displayed data; Matching the product information of the recorded product with the product information of the corresponding consulting product to determine whether the product information of the recorded product completely matches the product information of the consulting product; if the product information of the recorded product does not completely match the product information of the consulting product, determining that the displayed data is inaccurate and regenerating the displayed data; If the product information of the recorded product completely matches the product information of the consulted product, then the parameter items that successfully match are recorded according to the product information of the recorded product to obtain the parameter item sequence; Comparing the order of parameter items of multiple recorded products to determine whether the order of parameter items of multiple recorded products is the same; if the order of parameter items of multiple recorded products is different, determining that the displayed data is inaccurate, and regenerating the displayed data; If the order of the parameter items of the multiple recorded products is the same, it is determined that the displayed data is accurate and the displayed data is output.

8. A system for generating combination recommendation solutions based on product difference analysis, characterized in that: The system is used to implement a method for generating a combination recommendation solution based on product difference analysis as described in any one of claims 1 to 7, comprising: a data storage module, a demand judgment module, a data analysis module, and a data verification module; The data storage module obtains the product information of each product of the user, and performs attribute judgment on the product information of each product to determine the attribute information of each parameter item in the product information; The demand judgment module obtains customer demand information, judges the customer demand information, clarifies the products that the customer needs to consult and the corresponding product parameters, and marks the products that need to be consulted as consulting products and the corresponding product parameters that need to be consulted as consulting items; The data analysis module matches the consulting product with the stored products to determine the product information of the consulting product; judges the variety category of the consulting product, the number of consultations and the number of consultation items of the corresponding consulting product to determine the consulting status; the consulting status includes single-product multi-parameter consultation, single-product full-parameter consultation, multi-product multi-parameter consultation and multi-product full-parameter consultation; if the consulting status is single-product full-parameter consultation, the product information of the consulting product is output; if the consulting status is single-product multi-parameter consultation, the product information of the consulting product is processed according to the attribute information of the consulting items and parameter items to obtain the final display data; if the consulting status is multi-product full-parameter consultation, the product information of multiple consulting products is integrated to obtain display data with mutual comparison; if the consulting status is multi-product multi-parameter consultation, the product information of multiple consulting products is extracted and summarized according to the attribute information of the consulting items and parameter items to obtain display data of multiple consulting products compared with each other on the consulting items; The data verification module verifies the display data and outputs the verified data.

Images