CN113689260B - Commodity searching method and device - Google Patents
Commodity searching method and device Download PDFInfo
- Publication number
- CN113689260B CN113689260B CN202110948456.0A CN202110948456A CN113689260B CN 113689260 B CN113689260 B CN 113689260B CN 202110948456 A CN202110948456 A CN 202110948456A CN 113689260 B CN113689260 B CN 113689260B
- Authority
- CN
- China
- Prior art keywords
- commodity
- candidate
- commodities
- recommended value
- category
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application discloses a commodity searching method and device, and relates to the technical field of data processing. The server can determine a commodity category and a recommended value for each of the candidate commodities associated with the search keyword, and can rank the plurality of candidate commodities based on the commodity category and the recommended value for the candidate commodity. Among the sorted multiple candidate commodities, the commodity categories of the first N candidate commodities are different from each other, and the recommended value is sequentially decreased. Therefore, not only can the alternative commodity with higher matching degree with the search keyword be provided, but also the richness of commodity category of the alternative commodity can be ensured, thereby effectively improving the reliability of the search result pushed by the server.
Description
Technical Field
The present disclosure relates to the field of data processing technologies, and in particular, to a method and an apparatus for searching for a commodity.
Background
The terminal is generally provided with a shopping application program. The user may enter a keyword in a search box of the shopping application, which can send the keyword to the background server. The background server may obtain a plurality of candidate products whose product names match the keyword from the product database, and may sort the plurality of candidate products based on the matching degree of the product names and the keyword. And then, the background server can push commodity names of the sorted multiple candidate commodities to the shopping application program.
However, the method for sorting the candidate products based on the matching degree of the product names and the keywords is low in reliability, and therefore the search results pushed by the background server are low in reliability.
Disclosure of Invention
The application provides a commodity searching method and device, which can solve the technical problem of low reliability of search results pushed by a background server in the related technology. The technical scheme is as follows:
in one aspect, a method for searching for a commodity is provided, the method comprising:
determining a plurality of candidate commodities associated with the search keywords based on the acquired search keywords of the commodities;
determining commodity category and recommended value of each candidate commodity, wherein the recommended value is used for indicating the recommended degree of the candidate commodity, and the recommended value is positively correlated with the recommended degree;
sorting the plurality of candidate commodities based on commodity categories and recommended values of each candidate commodity, wherein the commodity categories of the first N candidate commodities are different from each other and the recommended values are sequentially decreased, N is the total number of commodity categories to which the plurality of candidate commodities belong, and N is an integer greater than 1;
And sending the ordered commodity information of the plurality of candidate commodities.
In another aspect, a server is provided, the server comprising: a communication module and a processor;
the processor is configured to:
determining a plurality of candidate commodities associated with the search keywords based on the acquired search keywords of the commodities;
determining commodity category and recommended value of each candidate commodity, wherein the recommended value is used for indicating the recommended degree of the candidate commodity, and the recommended value is positively correlated with the recommended degree;
sorting the plurality of candidate commodities based on commodity categories and recommended values of each candidate commodity, wherein the commodity categories of the first N candidate commodities are different from each other and the recommended values are sequentially decreased, N is the total number of commodity categories to which the plurality of candidate commodities belong, and N is an integer greater than 1;
the communication module is used for sending the ordered commodity information of the plurality of candidate commodities.
Optionally, the processor is configured to:
determining the arrangement order of the candidate commodity with the highest recommended value in the plurality of candidate commodities as the first order;
circularly executing the first sorting operation until the rest of the candidate commodities do not have the candidate commodities with different commodity categories from the candidate commodities with the determined arrangement order, so as to obtain the candidate commodities with the arrangement order at the first N positions; wherein the first sorting operation includes:
And determining the ranking order of the candidate commodity with the highest recommended value as the next position of the ordered candidate commodity in the candidate commodities with different commodity categories from the ordered candidate commodity.
Optionally, the processor is further configured to, after obtaining the ranking orders of the first N candidate products, sequentially determine, according to the order of the product categories of the first N candidate products, the ranking order of the candidate product with the highest recommended value among the remaining candidate products in each product category as a position after the ordered candidate products.
Optionally, the processor is further configured to:
after the arrangement order of the first N candidate commodities is obtained, determining that the arrangement order of the candidate commodity with the highest recommended value in the rest candidate commodities is N+1;
and circularly executing a second sorting operation until the rest of the candidate commodities have no candidate commodities with different commodity categories from the candidate commodities after the Nth position, wherein the second sorting operation comprises the following steps:
and determining the ranking order of the candidate commodity with the highest recommendation value as one position after the ordered candidate commodity in the candidate commodities with different commodity categories from the N-th candidate commodity.
Optionally, the processor is configured to determine, for each of the candidate commodities, a recommended value of the candidate commodity based on the first matching degree, the second matching degree and/or the first sales rate of the candidate commodity;
wherein the recommended value is positively correlated with the first matching degree, the second matching degree and the first sales rate; the first matching degree is the matching degree of the commodity name of the commodity candidate and the search keyword, the second matching degree is the matching degree of the standard class of the commodity candidate and the search keyword, and the first sales rate is determined according to the ratio of the sales volume of the commodity candidate to the total volume of the commodity candidate.
Optionally, the processor is configured to:
determining an initial recommended value of the alternative commodity based on the first matching degree, the second matching degree and/or the first sales rate of the alternative commodity;
updating the initial recommended value based on a second sales rate of the commodity of the target commodity category to obtain a recommended value of the alternative commodity, wherein the recommended value of the alternative commodity is positively correlated with the initial recommended value and the second sales rate;
the target commodity category is the commodity category of the alternative commodity, and the second sales rate is determined according to the ratio of the total sales volume of the commodity of the target commodity category to the total volume of the commodity of the target commodity category.
Optionally, the processor is configured to input commodity information of each candidate commodity into a class identification model, so as to obtain a commodity class of the candidate commodity output by the class identification model.
In yet another aspect, a server is provided, the server comprising: the system comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor realizes the commodity recommendation method according to the aspect when executing the computer program.
In still another aspect, there is provided a computer-readable storage medium having stored therein a computer program loaded by a processor and executing the search method of the commodity provided in the above aspect.
In yet another aspect, there is provided a computer program product containing instructions which, when run on a computer, cause the computer to perform the method of searching for merchandise provided in the above aspect.
The beneficial effects that this application provided technical scheme brought include at least:
the application provides a commodity searching method and device. The server can determine a commodity category and a recommended value for each of the candidate commodities associated with the search keyword, and can rank the plurality of candidate commodities based on the commodity category and the recommended value for the candidate commodity. Among the sorted multiple candidate commodities, the commodity categories of the first N candidate commodities are different from each other, and the recommended value is sequentially decreased. Therefore, not only can the alternative commodity with higher matching degree with the search keyword be provided, but also the richness of commodity category of the alternative commodity can be ensured, thereby effectively improving the reliability of the search result pushed by the server.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic structural diagram of a search system for commodities according to an embodiment of the present application;
FIG. 2 is a flowchart of a method for searching for a commodity according to an embodiment of the present application;
FIG. 3 is a flowchart of another method for searching for merchandise according to an embodiment of the present application;
FIG. 4 is a schematic diagram of a search interface provided by an embodiment of the present application;
FIG. 5 is a schematic diagram of a search system for another commodity according to an embodiment of the present application;
FIG. 6 is a schematic diagram of a search system for merchandise according to an embodiment of the present application;
FIG. 7 is a schematic diagram of a search results interface provided by an embodiment of the present application;
fig. 8 is a schematic structural diagram of a server according to an embodiment of the present application.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present application more apparent, the embodiments of the present application will be described in further detail below with reference to the accompanying drawings.
Fig. 1 is a schematic structural diagram of a search system for commodities according to an embodiment of the present application. Referring to fig. 1, the system may include: a server 110 and a terminal 120. Wherein a wired or wireless communication connection may be established between the server 110 and the terminal 120.
Alternatively, the server 110 may be a server, or may be a server cluster formed by a plurality of servers, or may also be a cloud computing service center. The terminal 120 may be a mobile phone, a tablet computer, a notebook computer, a desktop computer, a vehicle-mounted terminal, an intelligent home device, or a wearable device. For example, referring to fig. 1, the terminal 120 may be a mobile phone.
It will be appreciated that the terminal 120 may have a shopping Application (APP) installed therein, and that the server 110 is a background server of the shopping APP.
The present application provides a method for searching for a commodity, which may be applied to a server, such as the server 110 in the system shown in fig. 1. Referring to fig. 2, the method includes:
step 101, determining a plurality of candidate commodities associated with the search keyword based on the obtained search keyword of the commodity.
When a user needs to search for a commodity in a shopping APP installed on a terminal, a search keyword of the commodity can be input in a search box of the shopping APP. The terminal can then send the search keyword to the server. After receiving the search keyword of the commodity sent by the terminal, the server can determine a plurality of candidate commodities associated with the search keyword based on the commodity information of the commodity stored by the server. Wherein the commodity information may include at least one of the following information: commodity name, standard class of commodity, commodity number, commodity class and description information of commodity, etc. The descriptive information may be used to describe characteristics of the article, which may include: the color, size, shape, taste, flavor, use, place of origin, etc. of the commodity.
In the embodiment of the present application, the association of the commodity with the search keyword may refer to: the commodity information of the commodity contains part or all of the content of the search keyword.
Step 102, determining commodity category and recommended value of each candidate commodity.
Wherein the commodity category of each candidate commodity may refer to a standard category (e.g., a primary standard category, a secondary standard category, or a tertiary standard category) of the candidate commodity, and the server may determine the standard category from commodity information of the candidate commodity. Alternatively, the commodity category may be obtained by processing part or all of commodity information of the candidate commodity by the server using a category identification model.
The recommended value of the commodity is used for indicating the recommended degree of the candidate commodity, and the height of the recommended value is positively correlated with the height of the recommended degree. That is, the higher the degree of recommendation of the candidate commodity, the higher the recommendation value of the candidate commodity. In the embodiment of the application, the server may calculate the recommended value of each candidate commodity by adopting a prestored recommended value algorithm.
And step 103, sorting the plurality of candidate commodities based on commodity category and recommended value of each candidate commodity.
If N is the total number of commodity categories to which the plurality of candidate commodities belong, the commodity categories of the first N candidate commodities are different from each other in the plurality of ordered candidate commodities, and the recommended value is sequentially decreased. Wherein N is an integer greater than 1.
And 104, sending commodity information of the sorted multiple candidate commodities.
The server can send the ordered commodity information of the plurality of candidate commodities to the terminal as a search result, so that the terminal can display the commodity information of the plurality of candidate commodities in sequence.
In summary, the embodiment of the application provides a method for searching for a commodity. The server can determine a commodity category and a recommended value for each of the candidate commodities associated with the search keyword, and can rank the plurality of candidate commodities based on the commodity category and the recommended value for the candidate commodity. Among the sorted multiple candidate commodities, the commodity categories of the first N candidate commodities are different from each other, and the recommended value is sequentially decreased. Therefore, not only can the alternative commodity with higher matching degree with the search keyword be provided, but also the richness of commodity category of the alternative commodity can be ensured, thereby effectively improving the reliability of the search result pushed by the server.
The present example provides another method for searching for merchandise, which may be applied to a server, such as server 110 in the system shown in fig. 1. Referring to fig. 3, the method includes:
step 201, determining a plurality of candidate commodities associated with the search keyword based on the obtained search keyword of the commodity.
After receiving the search keyword of the commodity sent by the terminal, the server can determine a plurality of candidate commodities associated with the search keyword based on the commodity information of the commodity stored by the server. Wherein, the association of the commodity with the search keyword may refer to: the commodity information of the commodity contains part or all of the content of the search keyword. That is, if the product information of a certain product has an intersection with the search keyword, the product can be considered to be associated with the search keyword. For example, the server can obtain commodity information of the commodity from the commodity database and match with the search keyword.
In the embodiment of the present application, the commodity information of each commodity may include at least one of the following information: commodity name, standard class of commodity, commodity number and description information of commodity, etc. The descriptive information may be used to describe characteristics of the article, which may include: the color, size, shape, taste, flavor, use, place of origin, etc. of the commodity.
The standard categories of merchandise may include a plurality of different levels of categories, for example, a primary standard category, a secondary standard category, a tertiary standard category, and the like. Wherein the secondary standard class belongs to the subdivision class of the primary standard class, and the tertiary standard class belongs to the subdivision class of the secondary standard class. Also, the standard categories of merchandise may be manually configured by the merchant or may be automatically configured by the server.
The commodity number of the commodity is a character string capable of uniquely identifying the commodity, and for example, the commodity number may be a fixed-length number, or a combination of a number and a letter.
By way of example, assume that commodity information of a commodity stored in a server includes: commodity number, commodity name, description information, primary standard class, secondary standard class, and commodity class. Referring to fig. 4, fig. 4 is a schematic diagram of an interface with a search box displayed according to an embodiment of the present application. As shown in fig. 4, if the user inputs "onion" in the search box of the shopping APP installed at the terminal, the search keyword that the server can receive is "onion", and as shown in table 1, the plurality of candidate products determined by the server to be associated with the search keyword may include: the alternative commodity with commodity numbers 10001 to 10007. The commodity name of each candidate commodity contains a search keyword of 'onion'.
With continued reference to Table 1, the trade name of the alternative commodity with commodity number 10001 is "onion", the descriptive information is "about 240g-250 g/serving", the primary standard class is "fresh vegetables", the secondary standard class is "onion ginger seasoning", and the commodity class is "onion".
TABLE 1
Step 202, inputting commodity information of each candidate commodity into a class identification model to obtain the commodity class of the candidate commodity output by the class identification model.
In the embodiment of the application, a category identification model for identifying the commodity category of the commodity is prestored in the server, and the category identification model can be trained by adopting a machine learning method. After determining the candidate commodities, the server can input commodity information of each candidate commodity into the class identification model respectively to obtain commodity classes of the candidate commodities output by the class identification model. The server determines the commodity category of the candidate commodity through the pre-stored category identification model, so that the efficiency and accuracy in determining the commodity category can be improved.
It should be understood that the commodity category of the commodity output by the class identification model is the next level category (or may be referred to as sub-category) of the standard category of the commodity, that is, the commodity category of the commodity output by the class identification model is obtained by further dividing the standard category. For example, the commodity category of the commodity output by the class identification model is a sub-category of the secondary standard category.
It should also be appreciated that the server may input some or all of the information in the merchandise information for each candidate merchandise to the category identification model to obtain the merchandise category for that candidate merchandise. For example, the server may input only the commodity name and standard class of the candidate commodity to the class identification model.
Alternatively, the class identification model may include: labeling the sub-model and identifying the sub-model. The server can input commodity information of the alternative commodity into the labeling sub-model to obtain a labeling result output by the labeling sub-model. And then, the labeling result can be used as a candidate category to be input into the recognition sub-model, so that the commodity category of the candidate commodity output by the recognition sub-model is obtained. The labeling sub-model and the recognition sub-model can be machine learning models. The labeling sub-model may be a long short-term memory (LSTM) and conditional random field (conditional random field, CRF) based model. The recognition sub-model may be a model based on a convolutional neural network (convolutional neural network, CNN) and a flexible transmission value maximum function (softmax).
For example, the server may first combine the commodity name and standard class of the candidate commodity into a text, and convert the words in the text into word vectors to be input to the labeling sub-model. The labeling sub-model can label the input word vectors and output labeling results, and the number of the labeling results output by the labeling sub-model can be 0, 1 or more. Wherein each labeling result may be a keyword extracted from the commodity name and the standard category. And then, the server can take the labeling result and the standard category of the candidate commodity as candidate categories, respectively convert each candidate category into word vectors and input the word vectors into the recognition submodel. The identification sub-model can then output the commodity category of the candidate commodity.
Referring to table 1, for the candidate commodity of commodity number 10001, the server inputs commodity information of the candidate commodity into the class identification model, and then can determine that the commodity class of the candidate commodity is "onion". For the candidate commodity with commodity number 10002, after the server inputs commodity information of the candidate commodity into the class identification model, the commodity class of the candidate commodity can be determined as "green onion". For the candidate commodity with commodity number 10006, after the server inputs commodity information of the candidate commodity into the class identification model, the commodity class of the candidate commodity can be determined as "chives".
Optionally, fig. 5 is a schematic structural diagram of another commodity searching system according to an embodiment of the present application. As shown in fig. 5, the server may store the commodity information of the newly added commodity sent by the terminal in the commodity database each time the newly added commodity is put in storage. The server may input the commodity information of the newly added commodity into the class identification model to obtain the commodity class of the newly added commodity. The server may then store the commodity category of the newly added commodity as part of the commodity information, for example, in a commodity database. Accordingly, after the above step 201, the server may directly determine the commodity category of the candidate commodity from the commodity information of the candidate commodity stored in advance, without determining the commodity category of the candidate commodity through the category identification model.
Step 203, determining an initial recommended value of the candidate commodity based on the first matching degree, the second matching degree and/or the first sales rate of the candidate commodity.
In the embodiment of the application, the server may determine the initial recommended value of the candidate commodity based on at least one parameter of the first matching degree, the second matching degree and the first sales rate of the candidate commodity. And the initial recommended value of the candidate commodity is positively correlated with the first matching degree, the second matching degree and the first sales rate of the candidate commodity. I.e. the higher the first degree of matching of the candidate commodity, the higher the initial recommendation value of the candidate commodity. The higher the second degree of matching of the candidate commodity, the higher the initial recommendation value of the candidate commodity. The higher the first sales rate of the alternative merchandise, the higher the initial recommendation value of the alternative merchandise.
The first matching degree is the matching degree of commodity names of the candidate commodities and the search keywords. The second matching degree is the matching degree of the standard category of the candidate commodity and the search keyword. The first sales rate is determined from a ratio of sales of the candidate commodity to a total of the candidate commodity, and the first sales rate is positively correlated with the ratio. For example, the first sales rate may be equal to the ratio.
Alternatively, the server may directly determine the first matching degree, the second matching degree, or the first sales rate as the initial recommended value of the candidate commodity. Or the server may perform weighted summation on at least two parameters of the first matching degree, the second matching degree and the first sales rate to obtain an initial recommended value of the alternative commodity. Wherein the weight of each parameter may be pre-stored in the server.
It is understood that the matching degree of the commodity name (or standard class) and the search keyword may refer to the similarity of the two, that is, the more the same content the two contain, the higher the matching degree. The first sales rate may be determined based on a ratio of sales to total of the candidate good over the target period. The target period may be the last month, or the last three months, etc. The first matching degree and the second matching degree can reflect the similarity between the candidate commodity and the search keyword, and the first sales rate can reflect the popularity of the candidate commodity, so that the initial recommendation value of the determined candidate commodity is accurate based on the first matching degree, the second matching degree and/or the first sales rate.
Fig. 6 is a schematic structural diagram of a search system for a commodity according to another embodiment of the present application. As shown in fig. 6, the server further includes a ranking module that is capable of determining an initial recommendation value for the candidate commodity based on the first degree of matching, the second degree of matching, and/or the first rate of sales for the candidate commodity.
And 204, updating the initial recommended value based on the second sales rate of the commodity of the target commodity category to obtain the recommended value of the alternative commodity.
In this embodiment of the present application, after determining the initial recommended value of each candidate commodity, the server may further update the initial recommended value based on the second sales rate of the commodity of the target commodity category, to obtain the final recommended value of the candidate commodity. The recommended value of the commodity is used for indicating the recommended degree of the candidate commodity, and the height of the recommended value is positively correlated with the height of the recommended degree. That is, the higher the degree of recommendation of the candidate commodity, the higher the recommendation value of the candidate commodity. For example, the ranking module in the server can update the initial recommendation value based on the second sales rate to obtain a recommendation value for the alternative merchandise.
Wherein the target commodity category refers to the commodity category of the alternative commodity. The second sales rate is determined based on a ratio of a total sales volume of the items of the target item category to a total volume of the items of the target item category. That is, the second sales rate is determined according to the ratio of the total sales of all the commodities of the commodity category as the target commodity category to the total amount of all the commodities of the commodity category as the target commodity category. Because the second sales rate can reflect the overall popularity of the commodity of the target commodity category, updating the initial recommendation value based on the second sales rate can ensure that the recommendation value obtained after updating is more accurate.
It will be appreciated that the recommended value for the alternative good is positively correlated with both the initial recommended value and the second sales rate. I.e., the higher the initial recommendation value for the candidate commodity, the higher the recommendation value for the candidate commodity. The higher the second sales rate of the alternative merchandise, the higher the recommended value of the alternative merchandise. For example, assuming that the initial recommended value of a certain candidate commodity is V0 and the second sales rate of the commodity of the target commodity category is W, the recommended value V1 of the candidate commodity may satisfy: v1= (v0+k×w) ×p. Wherein the values of V0 and W can be in the range of [0,1]. K is a coefficient (or may be referred to as a weight) of the second sales rate P, and K may be a fixed value stored in advance in the server, and its value range may be [0,1]. P may be a fixed value that is not zero stored in advance in the server, for example, P may be equal to 1 or may be equal to 100.
For example, after calculating the recommended values of the candidate products in table 1 by using the methods shown in the above steps 203 and 204, the server may obtain the calculation results shown in table 2. Referring to table 2, assuming that the upper limit of the recommended value is 100, the recommended value of the candidate commodity with the commodity number 10001 is 86, the recommended value of the candidate commodity with the commodity number 10002 is 84, and the recommended value of the candidate commodity with the commodity number 10003 is 80.
TABLE 2
Step 205, determining the ranking order of the candidate commodity with the highest recommended value in the plurality of candidate commodities as the first ranking order.
The server can determine, based on the determined recommended value of each candidate commodity, the ranking of the candidate commodity with the highest recommended value as the first ranking, that is, determine the ranking of the candidate commodity with the highest recommended value as 1. It should be understood that the more forward the order of arrangement of the candidate commodity, the more forward the commodity information of the candidate commodity is displayed in the terminal.
For example, referring to table 2, since the recommended value of the candidate commodity with the commodity number 10001 is highest among the 7 candidate commodities shown in table 2, the server may determine that the ranking order of the candidate commodity with the commodity number 10001 is the first order.
And 206, circularly executing the first sorting operation until the rest of the candidate commodities have no candidate commodities with different commodity categories from the candidate commodities with the determined arrangement order, and obtaining the candidate commodities with the arrangement order at the first N positions.
Wherein the first sorting operation includes: and determining the ranking order of the candidate commodity with the highest recommended value as the next position of the ordered candidate commodity in the candidate commodities with different commodity categories from the ordered candidate commodity. That is, the server can sort, by the first sort operation, the candidate products, among the remaining candidate products, whose product categories are different from the candidate products of the determined sort order, after determining that the sort order of the products of the highest recommended value among the candidate products is the first order. When the server determines that there is no candidate commodity of the commodity category different from the candidate commodity of the determined ranking order among the remaining candidate commodities, the first ranking operation is stopped. Based on the above, the server can determine the candidate commodities whose arrangement order is in the first N positions, and the commodity categories of the N candidate commodities in the first N positions are different from each other, and the recommended value is sequentially decreased. Therefore, the server can be ensured to provide the alternative commodity with high matching degree with the search keyword, and the richness of commodity category of the alternative commodity can be ensured. Where N is the total number of commodity categories for the plurality of candidate commodities.
As illustrated in table 2, the total number of commodity categories for 7 candidate commodities with commodity numbers 10001 to 10007 is 4, i.e., n=4. After determining that the ranking order of the candidate commodities with commodity number 10001 is the first order, the server can start to circularly execute the first sorting operation. Since the commodity category of the candidate commodity with the commodity number 10001 is "onion", the server can determine the ranking order of the candidate commodity with the highest recommended value (i.e., the candidate commodity with the commodity number 10002) among the candidate commodities with the commodity categories different from "onion" as shown in table 3 as the second ranking.
At this time, the commodity categories of the sorted candidate commodities include "onion" and "green onion". The server can further determine the ranking order of the candidate product with the highest recommended value (i.e., the candidate product with the product number 10006) as the third order among the candidate products with the product categories different from "onion" and "welsh onion".
At this time, the commodity categories of the sorted candidate commodities include "onion", "welsh onion", and "chives". The server may also determine, as the fourth rank, the ranking rank of the candidate commodity having the highest recommended value (i.e., the candidate commodity having the commodity number 10007) among the candidate commodities having commodity categories different from "onion", "welsh onion", and "chive".
To this end, among the remaining candidate products (i.e., candidate products having product numbers 10003, 10004, and 10005), there are no candidate products of different product categories from those of the candidate products of the determined ranking order. The server stops performing the first sorting operation to obtain the candidate items with the ranking order of the first 4 bits as shown in table 3.
TABLE 3 Table 3
Step 207, determining that the ranking order of the candidate products with the highest recommended value in the rest candidate products is n+1.
After obtaining the candidate commodities with the ranking order at the first N digits, the server may determine the ranking order of the candidate commodity with the highest recommended value in the remaining candidate commodities as n+1 digits.
For example, as shown in table 3, after obtaining the candidate products with the ranking order of the first 4, the server determines the ranking order of the candidate product with the highest recommended value, that is, the candidate product with the product number of 10003, in the remaining candidate products as the 5 th bit.
And step 208, circularly executing the second sorting operation until the rest of the candidate commodities have no candidate commodities with different commodity categories from the candidate commodities after the Nth position.
Wherein the second sorting operation includes: and determining the ranking order of the candidate commodity with the highest recommended value as one position after the ordered candidate commodity in the candidate commodities with different commodity categories from the N-th candidate commodity. That is, after determining that the ranking order is the first N-bit candidate commodity and determining that the ranking order of the candidate commodity with the highest recommended value in the remaining candidate commodities is n+1-bit, the server can rank the candidate commodities with different commodity categories from the candidate commodities after the nth bit among the remaining candidate commodities through the second ranking operation. And stopping executing the second sorting operation when the server determines that the rest of the candidate commodities do not have the candidate commodities with the commodity category different from that of the candidate commodities after the Nth position.
For example, after determining the candidate product with the ranking order of n+1, the server may determine the ranking order of the candidate product with the highest recommended value as the n+2-th position from among the remaining candidate products, which are different from the n+1-th candidate product in the product category. Then, the server may continue to determine, as the n+3-th position, the ranking order of the candidate product having the highest recommended value among the remaining candidate products, which are different from the product categories of the n+1-th and n+2-th candidate products. And the like, until the rest of the candidate commodities have no candidate commodity with different commodity categories from the candidate commodities after the Nth position.
For example, referring to table 3, after determining that the ranking order of the candidate commodity with the commodity number 10003 is the 5 th order, the server may start to perform the second sorting operation in a loop. Since the commodity category of the candidate commodity with the ranking order of 5 (i.e., the candidate commodity with the commodity number of 10003) is "green onion", the server can determine the ranking order of the candidate commodity with the highest recommendation value (i.e., the candidate commodity with the commodity number of 10004) as the 6 th position, as shown in table 3, from among the remaining candidate commodities, the commodity category is different from "green onion". At this time, the commodity categories of the sorted candidate commodities after the 4 th place include "green Chinese onion" and "chives". The server may stop the second sorting operation because there is no alternative commodity of the remaining alternative commodities that is different from both the commodity categories of "green Chinese onion" and "chive".
It should be appreciated that after step 208 described above, the server may further continue to perform multiple rounds of sorting operations, where each round of sorting operations may sort a plurality of candidate products having different product categories from each other among the remaining candidate products in order of the recommended value from high to low. Based on the second sorting operation and the subsequently executed multi-round sorting operation, the commodity category of the alternative commodity positioned behind the Nth position can be ensured to be rich.
For example, assuming that the number of ordered candidate products is n+m (M is a positive integer not greater than N) after the above step 208, the server may determine the ranking order of the candidate products having the highest recommended value among the remaining candidate products as n+m+1 bits. Then, the server can circularly execute the third sorting operation until no alternative commodity with the commodity category different from the alternative commodity after the n+M bit exists in the rest alternative commodities. Wherein the third sorting operation includes: and determining the ranking order of the candidate commodity with the highest recommended value as one position after the ordered candidate commodity in the candidate commodities with different commodity categories from the candidate commodity with the N+M position.
After the step 206, as an alternative implementation manner, the server may further determine, directly according to the order of the commodity categories of the first N candidate commodities, the ranking order of the candidate commodity with the highest recommendation value in the remaining candidate commodities in each commodity category as the next one after the ordered candidate commodity. In the sorting process, if the remaining candidate commodities are of the same category, the server can sort the remaining candidate commodities according to the order of the recommended value from high to low.
For example, referring to table 3, the server may determine the ranking order of the candidate commodity having the commodity category "green onion" and the highest recommended value (i.e., the candidate commodity having the commodity number 10003) as the 5 th position, among the remaining candidate commodities, in order of the first 4 candidate commodities, that is, in order of the commodity categories "green onion", "chives", and "green onion". Then, the ranking order of the candidate product having the highest recommended value (i.e., the candidate product having the product number 10004) among the remaining candidate products, the product category of which is "chives", may be determined as the 6 th position. Finally, the server may determine the ranking order of the remaining candidate products with the product number 10005 as the 7 th bit.
As another alternative implementation manner, the server may also directly arrange the remaining candidate commodities in the order of the recommended value from high to low.
For example, referring to table 3, after obtaining the candidate products with the ranking order of the first 4 bits, the server directly sorts the remaining 3 candidate products in order of recommended value from high to low. That is, the order of the candidate commodity with the commodity number 10003 is the 5 th position, the order of the candidate commodity with the commodity number 10004 is the 6 th position, and the order of the candidate commodity with the commodity number 10005 is the 7 th position.
Step 209, sending commodity information of the sorted multiple candidate commodities.
After the server performs the sorting operation on the plurality of candidate commodities, the sorted commodity information of the plurality of candidate commodities can be sent to the terminal as a search result, so that the terminal can display the commodity information of the plurality of candidate commodities in sequence. That is, the terminal can display the commodity information of the plurality of candidate commodities according to the arrangement order determined by the server. For example, referring to fig. 6, after the sorting module in the server performs the sorting operation by the method shown in the above steps 205 to 208, the product information of the sorted multiple candidate products can be sent to the terminal as a search result. Wherein the commodity information of each candidate commodity may include at least one of a commodity name, a standard class of the commodity, a commodity number, and description information of the commodity.
For example, the terminal may display the search results as shown in table 4. Referring to table 4, commodity information of 7 candidate commodities is included in the search result, and commodity information of each candidate commodity includes a commodity number, a commodity name, and description information of the commodity. The 7 candidate products are arranged in the order of product numbers 10001, 10002, 10003, 10004, 10005, 10006, and 10007. Referring to fig. 7, fig. 7 is a schematic diagram of an interface displaying search results according to an embodiment of the present application. As shown in fig. 7, the terminal may display the commodity information of the plurality of candidate commodities in the arrangement order.
TABLE 4 Table 4
It should be understood that the sequence of the steps of the method for searching the commodity provided in the embodiments of the present application may be appropriately adjusted, and the steps may also be increased or decreased accordingly according to the situation. For example, step 204 may be omitted as the case may be, i.e., the server may directly determine the initial recommendation value as the final recommendation value for the candidate commodity without updating the initial recommendation value with the second sales rate. Alternatively, steps 207 and 208 may be omitted as the case may be, i.e., the server may also sort the remaining candidate products directly in the order of the recommended value from high to low. Any method that can be easily conceived by those skilled in the art within the technical scope of the present disclosure should be covered in the protection scope of the present application, and thus will not be repeated.
In summary, the embodiment of the application provides a method for searching for a commodity. The server can determine a commodity category and a recommended value for each of the candidate commodities associated with the search keyword, and can rank the plurality of candidate commodities based on the commodity category and the recommended value for the candidate commodity. Among the sorted multiple candidate commodities, the commodity categories of the first N candidate commodities are different from each other, and the recommended value is sequentially decreased. Therefore, not only can the alternative commodity with higher matching degree with the search keyword be provided, but also the richness of commodity category of the alternative commodity can be ensured, thereby effectively improving the reliability of the search result pushed by the server.
Fig. 8 is a schematic structural diagram of a server provided in an embodiment of the present application, where the server may perform the method for searching for a commodity performed by the server provided in the foregoing method embodiment. As shown in fig. 8, the server 110 includes: a processor 1101 and a communication module 1102.
The processor 1101 is configured to:
determining a plurality of candidate commodities associated with the search keyword based on the acquired search keyword of the commodity;
determining commodity category and recommended value of each candidate commodity, wherein the recommended value is used for indicating the recommended degree of the candidate commodity, and the height of the recommended value is positively correlated with the height of the recommended degree;
And sorting the plurality of candidate commodities based on commodity categories and recommended values of each candidate commodity, wherein the commodity categories of the first N candidate commodities are different from each other in the sorted plurality of candidate commodities, and the recommended values are sequentially decreased, N is the total number of commodity categories to which the plurality of candidate commodities belong, and N is an integer greater than 1.
The communication module 1102 is configured to send the ordered commodity information of the plurality of candidate commodities.
Optionally, the processor 1101 is configured to:
determining the arrangement order of the candidate commodity with the highest recommended value in the plurality of candidate commodities as the first order;
circularly executing the first sorting operation until the rest of the candidate commodities do not have the candidate commodities with different commodity categories from the candidate commodities with the determined arrangement order, so as to obtain the candidate commodities with the arrangement order at the first N positions; wherein the first sorting operation includes:
and determining the ranking order of the candidate commodity with the highest recommended value as the next position of the ordered candidate commodity in the candidate commodities with different commodity categories from the ordered candidate commodity.
Optionally, the processor 1101 is further configured to, after obtaining the ranking of the first N candidate products, sequentially determine, from the remaining candidate products of each product category, the ranking of the candidate product with the highest recommendation value as a position after the ordered candidate products, according to the order of product categories of the first N candidate products.
Optionally, the processor 1101 is further configured to:
after the arrangement order of the first N candidate commodities is obtained, determining that the arrangement order of the candidate commodity with the highest recommended value in the rest candidate commodities is N+1;
and circularly executing a second sorting operation until the rest of the candidate commodities have no candidate commodities with different commodity categories from the candidate commodities after the Nth position, wherein the second sorting operation comprises the following steps:
and determining the ranking order of the candidate commodity with the highest recommendation value as one position after the ordered candidate commodity in the candidate commodities with different commodity categories from the N-th candidate commodity.
Optionally, the processor 1101 is configured to determine, for each candidate commodity, a recommended value of the candidate commodity based on the first matching degree, the second matching degree and/or the first sales rate of the candidate commodity;
wherein the recommended value is positively correlated with the first matching degree, the second matching degree and the first sales rate; the first matching degree is the matching degree of the commodity name of the commodity candidate and the search keyword, the second matching degree is the matching degree of the standard class of the commodity candidate and the search keyword, and the first sales rate is determined according to the ratio of the sales volume of the commodity candidate to the total volume of the commodity candidate.
Optionally, the processor 1101 is configured to:
determining an initial recommended value of the alternative commodity based on the first matching degree, the second matching degree and/or the first sales rate of the alternative commodity;
updating the initial recommended value based on the second sales rate of the commodity of the target commodity category to obtain a recommended value of the alternative commodity, wherein the recommended value of the alternative commodity is positively correlated with the initial recommended value and the second sales rate;
the target commodity category is the commodity category of the alternative commodity, and the second sales rate is determined according to the ratio of the total sales volume of the commodity of the target commodity category to the total volume of the commodity of the target commodity category.
Optionally, the processor 1101 is configured to input the commodity information of each candidate commodity into the class identification model, so as to obtain the commodity class of the candidate commodity output by the class identification model.
In summary, the embodiment of the application provides a server. The server can determine a commodity category and a recommended value for each of the candidate commodities associated with the search keyword, and can rank the plurality of candidate commodities based on the commodity category and the recommended value for the candidate commodity. Among the sorted multiple candidate commodities, the commodity categories of the first N candidate commodities are different from each other, and the recommended value is sequentially decreased. Therefore, not only can the alternative commodity with higher matching degree with the search keyword be provided, but also the richness of commodity category of the alternative commodity can be ensured, thereby effectively improving the reliability of the search result pushed by the server.
The embodiment of the application provides a server, which may include a memory, a processor and a computer program stored on the memory and capable of running on the processor, where the processor implements the method for searching the commodity provided in the above embodiment when executing the computer program, for example, the method for searching the commodity shown in fig. 2 or fig. 3.
The present embodiment provides a computer-readable storage medium in which a computer program is stored, the computer program being loaded by a server and executing the search method for a commodity provided in the above embodiment, for example, the search method for a commodity shown in fig. 2 or fig. 3.
The embodiments of the present application also provide a computer program product containing instructions, which when executed on a computer, cause the computer to perform the method for searching for a commodity provided by the method embodiment described above, for example, the method for searching for a commodity shown in fig. 2 or fig. 3.
It will be understood by those skilled in the art that all or part of the steps for implementing the above embodiments may be implemented by hardware, or may be implemented by a program for instructing relevant hardware, where the program may be stored in a computer readable storage medium, and the storage medium may be a read-only memory, a magnetic disk or an optical disk, etc.
It should be understood that references herein to "and/or" means that there may be three relationships, e.g., a and/or B, which may mean: a exists alone, A and B exist together, and B exists alone. The character "/" generally indicates that the context-dependent object is an "or" relationship. Also, the terms "at least one" and "at least one" in this application mean one or more, and the term "plurality" in this application means two or more.
The terms "first," "second," and the like in this application are used to distinguish between identical or similar items that have substantially the same function and function, and it should be understood that there is no logical or chronological dependency between the "first," "second," and "nth" terms, nor is it limited to the number or order of execution.
The foregoing description of the exemplary embodiments of the present application is not intended to limit the invention to the particular embodiments disclosed, but on the contrary, the intention is to cover all modifications, equivalents, alternatives, and alternatives falling within the spirit and scope of the invention.
Claims (10)
1. A method of searching for a commodity, the method comprising:
determining a plurality of candidate commodities associated with the search keywords based on the acquired search keywords of the commodities;
Determining commodity category and recommended value of each candidate commodity, wherein the recommended value is used for indicating the recommended degree of the candidate commodity, and the recommended value is positively correlated with the recommended degree;
sorting the plurality of candidate commodities based on commodity categories and recommended values of each candidate commodity, wherein the commodity categories of the first N candidate commodities are different from each other and the recommended values are sequentially decreased, N is the total number of commodity categories to which the plurality of candidate commodities belong, and N is an integer greater than 1;
and sending the ordered commodity information of the plurality of candidate commodities.
2. The method of claim 1, wherein the ranking the plurality of candidate items based on the item category and the recommended value for each of the candidate items comprises:
determining the arrangement order of the candidate commodity with the highest recommended value in the plurality of candidate commodities as the first order;
circularly executing the first sorting operation until the rest of the candidate commodities do not have the candidate commodities with different commodity categories from the candidate commodities with the determined arrangement order, so as to obtain the candidate commodities with the arrangement order at the first N positions; wherein the first sorting operation includes:
And determining the ranking order of the candidate commodity with the highest recommended value as the next position of the ordered candidate commodity in the candidate commodities with different commodity categories from the ordered candidate commodity.
3. The method of claim 2, wherein after obtaining the ranking of the first N candidate items, the method further comprises:
and sequentially determining the ranking order of the candidate commodities with the highest recommended value in the rest candidate commodities in each commodity category as one position after the ordered candidate commodities according to the commodity category order of the first N candidate commodities.
4. The method of claim 2, wherein after obtaining the ranking of the first N candidate items, the method further comprises:
determining that the ranking order of the candidate commodity with the highest recommended value in the rest candidate commodities is N+1;
and circularly executing a second sorting operation until the rest of the candidate commodities have no candidate commodities with different commodity categories from the candidate commodities after the Nth position, wherein the second sorting operation comprises the following steps:
and determining the ranking order of the candidate commodity with the highest recommendation value as one position after the ordered candidate commodity in the candidate commodities with different commodity categories from the N-th candidate commodity.
5. The method of any one of claims 1 to 4, wherein determining a recommended value for each of the candidate items comprises:
for each of the candidate commodities, determining a recommended value of the candidate commodity based on the first matching degree, the second matching degree and/or the first sales rate of the candidate commodity;
wherein the recommended value is positively correlated with the first matching degree, the second matching degree and the first sales rate; the first matching degree is the matching degree of the commodity name of the commodity candidate and the search keyword, the second matching degree is the matching degree of the standard class of the commodity candidate and the search keyword, and the first sales rate is determined according to the ratio of the sales volume of the commodity candidate to the total volume of the commodity candidate.
6. The method of claim 5, wherein the determining the recommended value for the candidate item based on the first degree of matching, the second degree of matching, and/or the first sales rate for the candidate item comprises:
determining an initial recommended value of the alternative commodity based on the first matching degree, the second matching degree and/or the first sales rate of the alternative commodity;
updating the initial recommended value based on a second sales rate of the commodity of the target commodity category to obtain a recommended value of the alternative commodity, wherein the recommended value of the alternative commodity is positively correlated with the initial recommended value and the second sales rate;
The target commodity category is the commodity category of the alternative commodity, and the second sales rate is determined according to the ratio of the total sales volume of the commodity of the target commodity category to the total volume of the commodity of the target commodity category.
7. The method of any one of claims 1 to 4, wherein said determining a commodity category for each of said candidate commodities comprises:
inputting the commodity information of each candidate commodity into a class identification model to obtain the commodity class of the candidate commodity output by the class identification model.
8. A server, the server comprising: a communication module and a processor;
the processor is configured to:
determining a plurality of candidate commodities associated with the search keywords based on the acquired search keywords of the commodities;
determining commodity category and recommended value of each candidate commodity, wherein the recommended value is used for indicating the recommended degree of the candidate commodity, and the recommended value is positively correlated with the recommended degree;
sorting the plurality of candidate commodities based on commodity categories and recommended values of each candidate commodity, wherein the commodity categories of the first N candidate commodities are different from each other and the recommended values are sequentially decreased, N is the total number of commodity categories to which the plurality of candidate commodities belong, and N is an integer greater than 1;
The communication module is used for sending the ordered commodity information of the plurality of candidate commodities.
9. The server of claim 8, wherein the processor is configured to:
determining the arrangement order of the candidate commodity with the highest recommended value in the plurality of candidate commodities as the first order;
circularly executing the first sorting operation until the rest of the candidate commodities do not have the candidate commodities with different commodity categories from the candidate commodities with the determined arrangement order, so as to obtain the candidate commodities with the arrangement order at the first N positions; wherein the first sorting operation includes:
and determining the ranking order of the candidate commodity with the highest recommended value as the next position of the ordered candidate commodity in the candidate commodities with different commodity categories from the ordered candidate commodity.
10. The server of claim 8, wherein the processor is further configured to:
after the arrangement order of the first N candidate commodities is obtained, the arrangement order of the candidate commodities with the highest recommended value in the rest candidate commodities in each commodity category is determined to be one position behind the ordered candidate commodities according to the commodity category sequence of the first N candidate commodities.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110948456.0A CN113689260B (en) | 2021-08-18 | 2021-08-18 | Commodity searching method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110948456.0A CN113689260B (en) | 2021-08-18 | 2021-08-18 | Commodity searching method and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113689260A CN113689260A (en) | 2021-11-23 |
CN113689260B true CN113689260B (en) | 2023-08-08 |
Family
ID=78580396
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110948456.0A Active CN113689260B (en) | 2021-08-18 | 2021-08-18 | Commodity searching method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113689260B (en) |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103092858A (en) * | 2011-11-01 | 2013-05-08 | 阿里巴巴集团控股有限公司 | Search method and search device |
CN108460631A (en) * | 2018-02-13 | 2018-08-28 | 口口相传(北京)网络技术有限公司 | The mixing method for pushing and device of diversification information |
US10102559B1 (en) * | 2014-09-30 | 2018-10-16 | Amazon Technologies, Inc. | Diversification of recommendations |
CN108804605A (en) * | 2018-05-29 | 2018-11-13 | 重庆大学 | A kind of recommendation method based on hierarchical structure |
CN109376307A (en) * | 2018-12-19 | 2019-02-22 | 百度在线网络技术(北京)有限公司 | Article recommended method, device and terminal |
CN110263244A (en) * | 2019-02-14 | 2019-09-20 | 腾讯科技(深圳)有限公司 | Content recommendation method, device, storage medium and computer equipment |
US10423999B1 (en) * | 2013-11-01 | 2019-09-24 | Richrelevance, Inc. | Performing personalized category-based product sorting |
CN110610395A (en) * | 2018-06-15 | 2019-12-24 | 阿里巴巴集团控股有限公司 | Commodity search processing method, commodity ordering method, device and equipment |
CN110969516A (en) * | 2019-12-25 | 2020-04-07 | 清华大学 | Commodity recommendation method and device |
CN112288517A (en) * | 2019-12-23 | 2021-01-29 | 北京来也网络科技有限公司 | Commodity recommendation method and device combining RPA and AI |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6963867B2 (en) * | 1999-12-08 | 2005-11-08 | A9.Com, Inc. | Search query processing to provide category-ranked presentation of search results |
-
2021
- 2021-08-18 CN CN202110948456.0A patent/CN113689260B/en active Active
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103092858A (en) * | 2011-11-01 | 2013-05-08 | 阿里巴巴集团控股有限公司 | Search method and search device |
US10423999B1 (en) * | 2013-11-01 | 2019-09-24 | Richrelevance, Inc. | Performing personalized category-based product sorting |
US10102559B1 (en) * | 2014-09-30 | 2018-10-16 | Amazon Technologies, Inc. | Diversification of recommendations |
CN108460631A (en) * | 2018-02-13 | 2018-08-28 | 口口相传(北京)网络技术有限公司 | The mixing method for pushing and device of diversification information |
CN108804605A (en) * | 2018-05-29 | 2018-11-13 | 重庆大学 | A kind of recommendation method based on hierarchical structure |
CN110610395A (en) * | 2018-06-15 | 2019-12-24 | 阿里巴巴集团控股有限公司 | Commodity search processing method, commodity ordering method, device and equipment |
CN109376307A (en) * | 2018-12-19 | 2019-02-22 | 百度在线网络技术(北京)有限公司 | Article recommended method, device and terminal |
CN110263244A (en) * | 2019-02-14 | 2019-09-20 | 腾讯科技(深圳)有限公司 | Content recommendation method, device, storage medium and computer equipment |
CN112288517A (en) * | 2019-12-23 | 2021-01-29 | 北京来也网络科技有限公司 | Commodity recommendation method and device combining RPA and AI |
CN110969516A (en) * | 2019-12-25 | 2020-04-07 | 清华大学 | Commodity recommendation method and device |
Non-Patent Citations (1)
Title |
---|
针对结构化商品数据的多样性搜索***的设计与实现;陈相如;《中国优秀硕士学位论文全文数据库信息科技辑》;第7卷;I138-1520 * |
Also Published As
Publication number | Publication date |
---|---|
CN113689260A (en) | 2021-11-23 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109933660B (en) | API information search method towards natural language form based on handout and website | |
CA3059929C (en) | Text searching method, apparatus, and non-transitory computer-readable storage medium | |
CN110427480B (en) | Intelligent personalized text recommendation method and device and computer readable storage medium | |
CN112395487B (en) | Information recommendation method and device, computer readable storage medium and electronic equipment | |
US20160140389A1 (en) | Information extraction supporting apparatus and method | |
CN111428007B (en) | Cross-platform based synchronous push feedback method | |
CN113627995A (en) | Commodity recommendation list updating method and device, equipment, medium and product thereof | |
CN114492669B (en) | Keyword recommendation model training method, recommendation device, equipment and medium | |
US9442901B2 (en) | Resembling character data search supporting method, resembling candidate extracting method, and resembling candidate extracting apparatus | |
CN113591881B (en) | Intention recognition method and device based on model fusion, electronic equipment and medium | |
CN111915414B (en) | Method and device for displaying target object sequence to target user | |
CN112328881A (en) | Article recommendation method and device, terminal device and storage medium | |
CN113689260B (en) | Commodity searching method and device | |
US11947589B2 (en) | Web-scale personalized visual search recommendation service | |
CN116881432A (en) | Text pushing method, text pushing device, electronic equipment and storage medium | |
CN108334522B (en) | Method for determining customs code, and method and system for determining type information | |
CN116644102A (en) | Intelligent investment object selection method, system terminal and computer readable storage medium | |
CN112328918B (en) | Commodity sorting method, computing device and computer-readable storage medium | |
CN114048383A (en) | Information recommendation method and device, electronic equipment and medium | |
CN113987026A (en) | Method, apparatus, device and storage medium for outputting information | |
CN113656393B (en) | Data processing method, device, electronic equipment and storage medium | |
CN113344480B (en) | Replacement material recommendation method and system | |
CN117953273A (en) | Modeling method of learning-free artificial intelligence AI balance | |
CN112529676A (en) | Method for carrying out tax code final-stage classification based on social-level word vectors | |
CN110781383A (en) | Method for determining comprehensive sequencing investigation index |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |