CN-116501755-B - Multi-word search implementation method, device, medium and equipment

Abstract

The application provides a multi-word search implementation method, a device, a medium and equipment, wherein the method comprises the steps of determining a plurality of query words received by a search box, wherein the plurality of query words at least comprise first query words and second query words which have different semantics and correspond to different search terms, responding to a confirmation instruction of a multi-word intelligent collocation mode, sending the plurality of query words to a server to request intelligent collocation search based on the plurality of query words, receiving at least one intelligent collocation search result returned by the server, determining a target first search term and a target second search term according to an intelligent collocation strategy representing dependency among the query words, obtaining the intelligent collocation search result, and displaying the at least one intelligent collocation search result. The application can improve the searching efficiency.

Inventors

• ZHANG PENG
• WU ZHENLE
• ZHENG LI

Assignees

• 拉扎斯网络科技（上海）有限公司

Dates

Publication Date

20260508

Application Date

20230426

Claims (20)

1. 1. The method for realizing the multi-word search is characterized by being used for a terminal side, and comprises the following steps: Determining a plurality of query words received by a search box, wherein the plurality of query words at least comprise a first query word and a second query word which have different semantics and correspond to different search terms; responding to a confirmation instruction of the intelligent collocation mode of multiple words, sending the multiple query words to a server, and requesting intelligent collocation search based on the multiple query words; Receiving at least one intelligent collocation search result returned by a server, wherein a target first search term and a target second search term are determined according to an intelligent collocation strategy representing the dependency relationship between query words, the intelligent collocation search result is obtained, the intelligent collocation strategy is a first strategy with priority of collocation with stores or a second strategy for combining collocation between stores is determined according to a combined influence factor, the combined influence factor comprises at least one item of visit rate factor, distribution factor, price factor and space-time factor, if a CTR pre-estimated model is trained, when multi-word search is carried out, an intelligent collocation scheme is determined according to the CTR pre-estimated model, and an intelligent collocation result is generated and transmitted to a user, the CTR pre-estimated model is obtained through training according to a third fusion feature constructed by utilizing cross features and second fusion features, the cross features are obtained after cross linking the first fusion features obtained according to user side features and commodity features with query word side features, commodity features and logistics features, the second fusion features are obtained according to the query word side features and the query word side features, and the query word sequence comprises query word sequences and query word sequences, and the query word sequences are the query features; And displaying the at least one intelligent collocation search result.
2. 2. The method of claim 1, wherein determining the target first search term and the target second search term according to the intelligent collocation policy characterizing the dependency between query terms comprises: Confirming the intelligent collocation strategy as a first strategy of collocation priority with shops; and according to the first strategy of collocation priority of the same shop, respectively determining a target first search term and a target second search term belonging to the same shop from at least one first search term corresponding to the first query term and at least one second search term corresponding to the second query term.
3. 3. The method of claim 1, wherein determining the target first search term and the target second search term according to the intelligent collocation policy characterizing the dependency between query terms comprises: Determining a second strategy of combination collocation among stores according to the combination influence factors; and determining a target first search term and a target second search term corresponding to the combination influence factor from at least one first search term corresponding to the first query term and at least one second search term corresponding to the second query term according to a second strategy of combination collocation among stores.
4. 4. The method of claim 3, wherein the determining the target first search term and the target second search term corresponding to the combined impact factor comprises: identifying the combined influencing factors; and determining corresponding target first search terms and target second search terms from the at least one first search term and the at least one second search term respectively according to at least one of the purchase rate factor, the distribution factor, the price factor and the space-time factor.
5. 5. The method of claim 4, wherein the determining the corresponding target first search term and target second search term from the at least one first search term and the at least one second search term, respectively, based on at least one of a visit rate factor, a distribution factor, a price factor, and a space-time factor, comprises: Determining at least one current combined influence factor according to the priority of the visit rate factor, the distribution factor, the price factor and the space-time factor, and respectively determining corresponding target first search term and target second search term from the at least one first search term and the at least one second search term according to the current at least one combined influence factor, or And responding to screening results of the purchase rate factor, the distribution factor, the price factor and the space-time factor, determining at least one currently screened combined influence factor, and respectively determining corresponding target first search terms and target second search terms from the at least one first search term and the at least one second search term according to the at least one currently combined influence factor.
6. 6. The method as recited in claim 1, further comprising: Responding to a switching instruction of a multi-word autonomous collocation mode, sending a plurality of query words to a server, and requesting autonomous collocation search based on the plurality of query words; Receiving at least one first search term returned by a server side aiming at the first query term and at least one second search term returned by the second query term, and determining autonomous collocation search results in response to a selection instruction of a target first search term and a target second search term; and displaying the autonomous collocation search result.
7. 7. The method of claim 6, further comprising, prior to the selection instruction responsive to the target first search term and the target second search term: a first selection page for displaying at least one first search term for a first query term, a second selection page for displaying at least one second search term for a second query term, and controlling the first selection page and the second selection page to be switched by clicking on the first query term and the second query term; And receiving a selection instruction for selecting a target first search term on the first selection page, and receiving a selection instruction for selecting a target second search term on the second selection page.
8. 8. The method of claim 7, wherein the controlling the first selection page and the second selection page to switch by clicking on a first query term and a second query term comprises: After the selection instruction of the target first search term is determined to be received, when a second query word is clicked, switching to the second selection page to display at least one second search term, wherein a combination influence factor of the target first search term is determined, and the ranking of the at least one second search term on the second selection page is determined according to the combination influence factor.
9. 9. The method of claim 8, wherein the combined impact factor comprises at least one of a co-shop factor, a visit rate factor, a distribution factor, a price factor; The method further includes prompting the combined impact factor when the second selected page rank displays the at least one second search term.
10. 10. The method of any of claims 1-9, further comprising, after said determining a plurality of query terms received by a search box: identifying special characters which are not characters in the search box; Judging whether the special character is a preset multi-word distinguishing character, if so, distinguishing each query word by the multi-word distinguishing character to obtain the first query word and the second query word.
11. 11. The method for realizing the multi-word search is characterized by being used for a server side, and comprises the following steps: Receiving a plurality of query words sent by a terminal, and confirming that the terminal requests multi-word intelligent collocation search, wherein the plurality of query words at least comprise a first query word and a second query word which have different semantics and correspond to different search terms; Determining at least one intelligent collocation search result and returning the at least one intelligent collocation search result to the terminal, wherein a target first search term and a target second search term are determined according to an intelligent collocation strategy representing the dependency relationship between query words, the intelligent collocation search result is obtained, the intelligent collocation strategy is a first strategy with preferential collocation with shops or a second strategy for combined collocation among shops according to combined influence factors, the combined influence factors comprise at least one of a purchase rate factor, a distribution factor, a price factor and a space-time factor, if a CTR pre-estimated model is trained, when multi-word search is carried out, an intelligent collocation scheme is determined according to the CTR pre-estimated model, and an intelligent collocation result is generated and transmitted to a user, the CTR pre-estimated model is obtained through training according to a third fusion feature constructed by utilizing cross features and commodity features, the cross features are obtained after the first fusion feature obtained according to the characteristics of a user end and the commodity features are cross-linked with characteristics of the query end features, commodity features and logistics features, the second fusion features are obtained according to the query words and the query word cross features and the query word cross features are obtained according to the query word and the query word cross features.
12. 12. The method of claim 11, wherein determining the target first search term and the target second search term according to the intelligent collocation policy characterizing the dependency between query terms comprises: Setting the intelligent collocation strategy as a first strategy which is matched with shops preferentially; And determining target first search terms and target second search terms belonging to the same store from at least one first search term recalled by the first query term and at least one second search term recalled by the second query term according to the first strategy of collocation priority of the same store.
13. 13. The method of claim 11, wherein determining the target first search term and the target second search term according to the intelligent collocation policy characterizing the dependency between query terms comprises: The intelligent collocation strategy is set as a second strategy for determining combination collocation among stores according to the combination influence factors; And determining a target first search term and a target second search term corresponding to the combination influence factor from at least one first search term recalled for the first query term and at least one second search term recalled for the second query term according to a second strategy of combination collocation among stores.
14. 14. The method of claim 13, wherein the determining the target first search term and the target second search term corresponding to the combined impact factor comprises: and determining corresponding target first search terms and target second search terms from the at least one first search term and the at least one second search term respectively according to at least one of the purchase rate factor, the distribution factor, the price factor and the space-time factor.
15. 15. The method of claim 14, wherein the determining the corresponding target first search term and target second search term from the at least one first search term and the at least one second search term, respectively, based on at least one of a purchase rate factor, a distribution factor, a price factor, and a space-time factor, comprises: Determining at least one current combined influence factor according to the priority of the visit rate factor, the distribution factor, the price factor and the space-time factor, and respectively determining corresponding target first search term and target second search term from the at least one first search term and the at least one second search term according to the current at least one combined influence factor, or And receiving screening results of the purchase rate factor, the distribution factor, the price factor and the space-time factor sent by the terminal, determining at least one current combined influence factor, and respectively determining corresponding target first search terms and target second search terms from the at least one first search term and the at least one second search term according to the at least one current combined influence factor.
16. 16. The method as recited in claim 11, further comprising: confirming that the terminal requests multi-word autonomous collocation search; Recall at least one first search term for the first query term, recall at least one second search term for the second query term, and return the at least one first search term and the at least one second search term to the terminal, control the terminal paging surface to display the at least one first search term and the at least one second search term, provide a selection button of a target first search term and a target second search term, and determine an autonomous collocation search result.
17. 17. The method of claim 16, wherein the control terminal paging plane presents the at least one first search term and the at least one second search term, comprising: Controlling the terminal to display a first selection page of a first search term for a first query term and a second selection page of a second search term for a second query term, and controlling the first selection page and the second selection page to be switched by clicking the first query term and the second query term; And controlling to receive a selection instruction for selecting a target first search term on the first selection page, and receiving a selection instruction for selecting a target second search term on the second selection page.
18. 18. The method of claim 17, wherein the controlling the first selection page and the second selection page to switch by clicking on a first query term and a second query term comprises: And after determining that the selection instruction of the target first search term is received, when a second query term is clicked, the terminal is controlled to switch to the second selection page to display at least one second search term, wherein a combination influence factor of the target first search term is determined, and the ordering of the at least one second search term on the second selection page is determined according to the combination influence factor.
19. 19. The method of claim 18, wherein the combined impact factor comprises at least one of a co-shop factor, a visit rate factor, a distribution factor, a price factor; the method further comprises the step of controlling the terminal to prompt the combined influence factor when the second selected page orders and displays the at least one second search term.
20. 20. The method of any of claims 11-19, further comprising, after receiving the plurality of query terms: identifying special characters which are not characters in the plurality of query words; Judging whether the special character is a preset multi-word distinguishing character, if so, distinguishing each query word by the multi-word distinguishing character to obtain the first query word and the second query word.

Description

Multi-word search implementation method, device, medium and equipment Technical Field The present application relates to the field of data processing technologies, and in particular, to a method, an apparatus, a medium, and a device for implementing multi-word search. Background When a user uses an APP or applet, a search function is often used to find a target object, for example, in an electronic market such as retail, restaurant, medicine, etc., by inputting a query word (query) in a search box, a target object (target commodity, target dish, target medicine), etc. The efficiency of browsing and determining effective search terms (items) by a user directly affects the conversion effect of the user, so how to improve the search efficiency is a technical problem to be solved. Disclosure of Invention In view of the above, the present application provides a method, apparatus, medium and device for implementing multi-word search, and the main purpose is to improve the search efficiency. According to one aspect of the application, a multi-word search implementation method is provided and used for a terminal side, the method comprises the steps of determining a plurality of query words received by a search box, wherein the plurality of query words at least comprise first query words and second query words which have different semantics and correspond to different search terms, sending the plurality of query words to a server side in response to a confirmation instruction of a multi-word intelligent collocation mode, requesting intelligent collocation search based on the plurality of query words, receiving at least one intelligent collocation search result returned by the server side, determining a target first search term and a target second search term according to an intelligent collocation strategy representing dependency among the query words, obtaining the intelligent collocation search result, and displaying the at least one intelligent collocation search result. According to one aspect of the application, a method for realizing multi-word searching is provided, which is used for a server side, and comprises the steps of receiving a plurality of query words sent by a terminal, confirming a terminal request to perform multi-word intelligent collocation searching, wherein the plurality of query words at least comprise a first query word and a second query word which have different semantics and correspond to different search terms, determining at least one intelligent collocation searching result, and returning the at least one intelligent collocation searching result to the terminal, wherein a target first search term and a target second search term are determined according to an intelligent collocation strategy which characterizes the dependency relationship among the query words, and the intelligent collocation searching result is obtained. According to one aspect of the application, a training method of a multi-word intelligent collocation prediction model is provided, which comprises the steps of obtaining sample data, carrying out feature engineering processing on the sample data to obtain input features of the prediction model, wherein the input features comprise at least one of user side features, commodity side features, query word side features, logistics side features and space-time side features, and learning a neural network structure with a dependency relationship among a plurality of query words according to at least one of the user side features, commodity features, query word side features, logistics features and space-time features to obtain a click through rate CTR prediction model. According to one aspect of the application, a multi-word search implementation device is provided, and the device is used for a terminal side and comprises a query word determining unit, an intelligent collocation request unit and an intelligent collocation confirmation unit, wherein the query word determining unit is used for determining a plurality of query words received by a search box, the plurality of query words at least comprise first query words and second query words which have different semantics and correspond to different search terms, the intelligent collocation request unit is used for responding to a confirmation instruction of a multi-word intelligent collocation mode, sending the plurality of query words to a server side and requesting intelligent collocation search based on the plurality of query words, the intelligent collocation confirmation unit is used for receiving at least one intelligent collocation search result returned by the server side, wherein a target first search term and a target second search term are determined according to an intelligent collocation strategy representing dependency among the query words, and the intelligent collocation search result is obtained, and the intelligent collocation display unit is used for displaying the at least one intelligent collocation sea