Customized pricing decision support system integrating recommendation algorithms

Abstract

This novel system relates to a customized pricing decision support system integrating a recommendation algorithm. It comprises a production database server, an analysis server, a results database, an application interface server, and a front-end pricing platform, all interconnected. The analysis server retrieves pricing data from the production database server and generates first recommendation results, second recommendation results, and preference category ranking results based on this data. It also displays to the target vendor cases where other vendors have participated in the bidding, newly listed cases of similar cases, and popular cases related to the target vendor's preference categories, thereby improving case search efficiency, pricing accuracy, and case matching efficiency.

Claims

1. A customized pricing decision support system integrating recommendation algorithms, comprising: a production database server for storing pricing data related to multiple vendors and multiple pricing cases; and an analysis server communicatively connected to the production database server, and comprising: A data synchronization unit is used to retrieve the quotation data from the production database server; a collaborative recommendation unit is used to generate a first recommendation result for a target vendor based on the historical quotation data of multiple vendors, wherein the first recommendation result includes candidate quotation cases from other vendors with quotation relevance to the target vendor who have participated in the quotation, but the target vendor has not yet participated in the quotation; a content recommendation unit is used to generate a second recommendation result based on the case similarity between the target vendor's historical quotation cases and the multiple candidate quotation cases; and a preference category ranking unit is used to generate a preference category ranking result based on the case category corresponding to the target vendor's valid quotation data; and a results database, communicatively connected to the analysis server, is used to store the first recommendation result, the second recommendation result, and the preference category ranking result; An application interface server is connected to the results database to obtain the first recommendation result, the second recommendation result, and the preference category ranking result from the results database; and a front-end pricing platform is connected to the application interface server and includes a display interface for displaying at least one of the following for the target vendor: first recommendation case information corresponding to the first recommendation result, second recommendation case information corresponding to the second recommendation result, and preference category case information corresponding to the preference category ranking result.

2. The customized quotation decision support system integrating recommendation algorithms as described in Request 1, wherein the quotation data includes vendor identification data, case identification data, quotation time data, case category data and quotation validity data, and the quotation validity data is used to indicate whether the corresponding quotation record is a valid quotation record with a filled unit price and submitted.

3. The customized pricing decision support system with integrated recommendation algorithm as described in claim 1, wherein the analysis server further includes a scheduling update unit for controlling the data synchronization time of the data synchronization unit and the update time of the first recommendation result, the second recommendation result and the preference category ranking result.

4. The customized pricing decision support system with integrated recommendation algorithm as described in claim 1, wherein the collaborative recommendation unit includes an association rule exploration module, which is used to perform association rule exploration on the pricing history of multiple vendors to establish the pricing relationship between vendors and cases.

5. The customized pricing decision support system with integrated recommendation algorithm as described in claim 4, wherein the association rule exploration module performs data exploration using the Apriori algorithm or other association rule analysis methods, and filters associated data based on at least one of support, confidence, or lift value, and classifies the pricing association relationship into multiple association levels based on the confidence value.

6. The customized quotation decision support system with integrated recommendation algorithm as described in claim 1, wherein the content recommendation unit includes a case attribute comparison module, which is used to calculate or determine the case similarity between the historical quotation cases of the target manufacturer and each candidate quotation case based on one or more case attributes, and the one or more case attributes include at least one of case category, product classification code, procurement category and case description.

7. The customized pricing decision support system with integrated recommendation algorithm as described in claim 1, wherein the preference category ranking unit includes a category statistics module, which is used to count the number of valid pricing records of the target manufacturer in each case category, and generate the preference category ranking result accordingly.

8. A customized pricing decision support system based on an integrated recommendation algorithm as described in claim 1, wherein the result database includes at least one of a relational database, a non-relational database, a full-text search database, or a search index database.

9. The customized pricing decision support system integrating recommendation algorithms as described in claim 1, wherein when the application interface server receives a page request related to the target vendor, it obtains at least one of the first recommendation result, the second recommendation result, and the preference category ranking result from the result database, and forms recommendation result feedback data to provide to the front-end pricing platform.

10. A customized pricing decision support system with an integrated recommendation algorithm as described in claim 1, wherein the analysis server further writes back at least a portion of the first recommendation result, the second recommendation result, the preference category ranking result, index data, or statistical data to the production database server.

11. A customized pricing decision support system with integrated recommendation algorithm as described in claim 1, wherein the display interface includes a first recommendation display area, a second recommendation display area, a third recommendation display area, and a preference category sorting area. The first recommendation display area is used to display first recommended case information corresponding to the first recommendation result. The second recommendation display area is used to display second recommended case information corresponding to the second recommendation result. The third recommendation display area is used to display popular case information related to the preference category sorting result. The preference category sorting area is used to display the preference category sorting result.

12. The customized pricing decision support system integrating recommendation algorithms as described in claim 1, wherein the front-end pricing platform further includes a notification module, which generates notification content based on the first recommendation result, the second recommendation result, or the preference category ranking result, and sends the notification content to the target manufacturer.

Images