CN111445302A - Commodity sorting method, system and device - Google Patents
Commodity sorting method, system and device Download PDFInfo
- Publication number
- CN111445302A CN111445302A CN201910045121.0A CN201910045121A CN111445302A CN 111445302 A CN111445302 A CN 111445302A CN 201910045121 A CN201910045121 A CN 201910045121A CN 111445302 A CN111445302 A CN 111445302A
- Authority
- CN
- China
- Prior art keywords
- sub
- commodity
- information
- displayed
- similarity
- 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.)
- Pending
Links
Images
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]
- G06Q30/0633—Lists, e.g. purchase orders, compilation or processing
Abstract
The embodiment of the invention provides a commodity ordering method, a system and a device, wherein the commodity ordering method comprises the following steps: acquiring commodity information of a commodity triggered by a user on a first commodity list page; after a page change instruction is received, updating a sorted list of commodities to be displayed on a second commodity list page according to the similarity of the commodity information and the page change instruction, wherein the second commodity list page is a commodity list page to which the first commodity list page jumps after the page change instruction; and displaying the commodities to be displayed on a second commodity list page according to the updated sorted list. The invention solves the technical problem that the prediction accuracy of the traditional commodity ordering method for the currently favored commodity of the user is poor.
Description
Technical Field
The present invention relates to the field of electronic commerce technologies, and in particular, to a method, a system, and an apparatus for sorting commodities.
Background
The commodity ordering transaction of the e-commerce industry occurs after a user makes a query, and a search engine recalls commodities to be displayed in front of an interface. Hundreds or thousands of items may be returned for a query by a user. How to realize an ideal commodity sequencing enables the commodity favored by the user to appear at the head end of the sequencing with the maximum probability is the key point of the search service of the e-commerce industry which is always dedicated to optimization.
The search service of the e-commerce industry is realized by three parts of cooperation: the system comprises an index file for storing commodity information, a search engine for recalling commodities according to queries and a commodity ordering model formed by a plurality of rules. On the premise that the index file, the search engine query sentence and the sequencing model are fixed, a commodity result page formed by a user through query is also fixed, and the personalized requirements of the user on commodities are ignored. At present, some e-commerce websites establish a user portrait for each user through user online shopping behavior data, and realize personalized sorting of user commodities based on the user portrait. Specifically, firstly, collecting behavior data of a series of online purchases such as user inquiry, clicking, car adding, ordering and the like in the online purchase of the user; then, determining the commodity purchased by the user through the collected behavior data, and updating the user portrait based on the commodity purchased by the user; next, the merchandise is ordered in the next online purchase based on the latest version of the user profile.
In the commodity sorting method, the user portrait based on which the current commodity sorting is carried out is constructed based on the previous online shopping data of the user, and the previous online shopping data of the user is obtained by taking previous search as a background, namely, the previous online shopping data of the user reflects online shopping behaviors of the user at the previous online shopping time and online shopping environment, and although the online shopping data has a prediction function with statistical significance on the popular commodity of the current user, the prediction accuracy is poor.
Disclosure of Invention
In view of the foregoing, an object of the present invention is to provide a method, a system and a device for product ranking to alleviate the technical problem of the conventional product ranking method that the prediction accuracy of the currently favored product of the user is poor.
In a first aspect, an embodiment of the present invention provides a method for sorting commodities, including:
acquiring commodity information of a commodity triggered by a user on a first commodity list page;
after a page change instruction is received, updating a ranked list of commodities to be displayed on a second commodity list page according to the similarity of the commodity information and the page change instruction, wherein the second commodity list page is a commodity list page to which the first commodity list page jumps after the page change instruction;
and displaying the commodities to be displayed on the second commodity list page according to the updated sorted list.
With reference to the first aspect, an embodiment of the present invention provides a first possible implementation manner of the first aspect, where updating the ordered list of the commodities to be displayed on the second commodity list page according to the similarity between the commodity information and the commodity information includes:
extracting a sub-information set of the triggered commodity from the commodity information, wherein the sub-information set at least comprises the following sub-information: picture information, title information, price information, label information;
calculating the similarity of the commodities to be displayed and each piece of sub information in the sub information set to obtain a first sub similarity, wherein the first sub similarity corresponds to the sub information one to one;
and updating the sorted list of the commodities to be displayed on the second commodity list page according to the first sub-similarity.
With reference to the first possible implementation manner of the first aspect, an embodiment of the present invention provides a second possible implementation manner of the first aspect, where a plurality of pieces of sub information in the sub information set are provided, and the updating of the ordered list of the commodities to be displayed on the second commodity list page according to the first sub similarity includes:
acquiring the first sub-similarity between the current to-be-displayed commodity and the plurality of pieces of sub-information to obtain a plurality of first sub-similarities of the current to-be-displayed commodity;
and updating the arrangement sequence of the current commodities to be displayed in the ordered list according to the plurality of first sub-similarities to obtain the updated ordered list.
With reference to the second possible implementation manner of the first aspect, an embodiment of the present invention provides a third possible implementation manner of the first aspect, where updating, according to a plurality of the first sub-similarities, an arrangement order of the current to-be-displayed commodity in the ordered list to obtain an updated ordered list includes:
acquiring weights corresponding to the sub information to obtain a plurality of sub weights, wherein the sub weights correspond to the sub information one to one;
calculating a weighted sum of a plurality of first sub-similarities based on the product of the sub-weight corresponding to each piece of the sub-information and the first sub-similarity;
and updating the arrangement sequence of the current commodities to be displayed in the ordered list according to the weighted sum.
With reference to the first possible implementation manner of the first aspect, an embodiment of the present invention provides a fourth possible implementation manner of the first aspect, where the sub information in the sub information set includes picture information, and the calculating a similarity between the to-be-displayed commodity and each piece of sub information in the sub information set to obtain a first sub similarity includes:
extracting local features from the picture of the commodity to be displayed based on a local feature detection algorithm to obtain a feature list formed by the SIFT features;
clustering the feature list by using a clustering algorithm, and determining a clustering center obtained by clustering as a standard feature of the feature list;
obtaining a theme with the maximum standard feature occurrence probability through a bag-of-words model, and determining the theme as the category of the commodity to be displayed;
and determining a first sub-similarity corresponding to the commodity to be displayed and the picture information according to the category of the commodity to be displayed and the similarity of the picture information.
In combination with the fourth possible implementation manner of the first aspect, the present invention provides a fifth possible implementation manner of the first aspect, wherein,
the local feature detection algorithm adopts an SIFT algorithm; and/or the presence of a gas in the gas,
the clustering algorithm adopts a K-means algorithm; and/or the presence of a gas in the gas,
the bag of words model adopts an L DA model.
With reference to the first aspect, an embodiment of the present invention provides a sixth possible implementation manner of the first aspect, where the number of triggered products is multiple, and the updating of the ordered list of the products to be displayed on the second product list page according to the similarity between the triggered products and the product information includes:
obtaining the similarity between the current to-be-displayed commodity and the plurality of triggered commodities to obtain a plurality of second sub-similarities of the current to-be-displayed commodity;
and updating the arrangement sequence of the current commodities to be displayed in the ordered list according to the sum of the second sub-similarities to obtain the updated ordered list.
In a second aspect, an embodiment of the present invention provides a commodity ordering system, including:
the acquisition module is used for acquiring commodity information of a commodity triggered by a user on a first commodity list page;
the updating module is used for updating the ordered list of the commodities to be displayed on a second commodity list page according to the similarity with the commodity information after receiving a page change instruction, wherein the second commodity list page is a commodity list page to which the first commodity list page jumps after the page change instruction;
and the display module is used for displaying the commodities to be displayed on the second commodity list page according to the updated sorted list.
With reference to the second aspect, an embodiment of the present invention provides a first possible implementation manner of the second aspect, where the update module includes:
an extracting unit, configured to extract a sub-information set of the triggered product from the product information, where the sub-information set at least includes one of the following pieces of sub-information: picture information, title information, price information, label information;
the calculation unit is used for calculating the similarity of the commodities to be displayed and each piece of sub information in the sub information set to obtain a first sub similarity, wherein the first sub similarity is in one-to-one correspondence with the sub information;
and the updating unit is used for updating the sorted list of the commodities to be displayed on the second commodity list page according to the first sub-similarity.
With reference to the first possible implementation manner of the second aspect, an embodiment of the present invention provides a second possible implementation manner of the second aspect, where a number of pieces of sub information in the sub information set is multiple, and the updating unit includes:
the obtaining subunit is configured to obtain the first sub-similarity between the current to-be-displayed commodity and the plurality of pieces of sub-information, so as to obtain a plurality of first sub-similarities of the current to-be-displayed commodity;
and the updating subunit is used for updating the arrangement sequence of the current commodities to be displayed in the ordered list according to the plurality of first sub-similarities so as to obtain the ordered list.
With reference to the second possible implementation manner of the second aspect, an embodiment of the present invention provides a third possible implementation manner of the second aspect, where the updating subunit is configured to:
acquiring weights corresponding to the sub information to obtain a plurality of sub weights, wherein the sub weights correspond to the sub information one to one;
calculating a weighted sum of a plurality of first sub-similarities based on the product of the sub-weight corresponding to each piece of the sub-information and the first sub-similarity;
and updating the arrangement sequence of the current commodities to be displayed in the ordered list according to the weighted sum.
With reference to the first possible implementation manner of the second aspect, an embodiment of the present invention provides a fourth possible implementation manner of the second aspect, where the sub information in the sub information set includes picture information, and the calculating unit is configured to:
extracting local features from the picture of the commodity to be displayed based on a local feature detection algorithm to obtain a feature list formed by the local features;
clustering the feature list by using a clustering algorithm, and determining a clustering center obtained by clustering as a standard feature of the feature list;
obtaining a theme with the maximum standard feature occurrence probability through a bag-of-words model, and determining the theme as the category of the commodity to be displayed;
and determining a first sub-similarity corresponding to the commodity to be displayed and the picture information according to the category of the commodity to be displayed and the similarity of the picture information.
In combination with the fourth possible implementation manner of the second aspect, the present invention provides a fifth possible implementation manner of the second aspect, wherein,
the local feature detection algorithm adopts an SIFT algorithm; and/or the presence of a gas in the gas,
the clustering algorithm adopts a K-means algorithm; and/or the presence of a gas in the gas,
the bag of words model adopts an L DA model.
With reference to the second aspect, an embodiment of the present invention provides a sixth possible implementation manner of the second aspect, where the number of triggered commodities is multiple, and the update module is configured to:
obtaining the similarity between the current to-be-displayed commodity and the plurality of triggered commodities to obtain a plurality of second sub-similarities of the current to-be-displayed commodity;
and updating the arrangement sequence of the current commodities to be displayed in the ordered list according to the sum of the second sub-similarities to obtain the updated ordered list.
In a third aspect, an embodiment of the present invention provides a computer-readable storage medium, where computer instructions are stored, and when the computer instructions are executed, the method for sorting commodities is implemented as in the first aspect.
In a fourth aspect, an embodiment of the present invention provides a commodity sorting apparatus, including:
a memory for storing computer instructions;
a processor coupled to the memory, the processor configured to perform a method of implementing the article ordering method as in the first aspect based on computer instructions stored by the memory.
The embodiment of the invention has the following beneficial effects:
the commodity ordering method provided by the embodiment of the invention comprises the following steps: acquiring commodity information of a commodity triggered by a user on a first commodity list page; after a page change instruction is received, updating a sorted list of commodities to be displayed on a second commodity list page according to the similarity of the commodity information and the page change instruction, wherein the second commodity list page is a commodity list page to which the first commodity list page jumps after the page change instruction; and displaying the commodities to be displayed on a second commodity list page according to the updated sorted list.
According to the commodity ordering method provided by the embodiment of the invention, after a user browses commodities on the first commodity list page, if the page is to be changed through a page changing instruction, commodities which are updated and arranged according to the similarity with the commodity information are obtained on the second commodity list page, so that the shopping behavior information generated on the first commodity list page by the user is fed back to the second commodity list page in real time, the purpose that the instant online shopping behavior of the user is fed back to the current commodity query transaction is achieved, the ordering of the commodities queried on the second commodity list page is optimized, the conformity between the commodity ordering of the second commodity list page and the search intention of the user is improved, and the technical problem that the accuracy of the traditional commodity ordering method for predicting the currently favored commodities of the user is poor is further solved.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and drawings.
In order to make the aforementioned and other objects, features and advantages of the present invention comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
The above and other objects, features and advantages of the present invention will become more apparent from the following description of the embodiments of the present invention with reference to the accompanying drawings, in which:
fig. 1 is a flowchart of a commodity sorting method according to an embodiment of the present invention;
fig. 2 is a flowchart of a method for updating an ordered list of commodities to be displayed on a second commodity list page according to similarity to commodity information according to an embodiment of the present invention;
fig. 3 is a flowchart of a method for calculating similarity between a product to be displayed and each piece of sub information in a sub information set to obtain a first sub similarity according to a first embodiment of the present invention;
fig. 4 is a flowchart of a method for updating an ordered list of commodities to be displayed on a second commodity list page according to a first sub-similarity according to an embodiment of the present invention;
fig. 5 is a flowchart of another method for updating the ordered list of the commodities to be displayed on the second commodity list page according to the similarity with the commodity information according to the first embodiment of the present invention;
fig. 6 is a block diagram illustrating a structure of a commodity sorting system according to a second embodiment of the present invention;
fig. 7 is a block diagram illustrating a structure of a commodity sorting apparatus according to a fourth embodiment of the present invention.
Icon: 100-an acquisition module; 200-an update module; 300-a display module; 701-a memory; 702-a processor; 703-input/output devices.
Detailed Description
Various embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Like elements in the various figures are denoted by the same or similar reference numerals. For purposes of clarity, the various features in the drawings are not necessarily drawn to scale.
The following detailed description of embodiments of the present invention is provided in connection with the accompanying drawings and examples.
At present, some e-commerce websites establish user portraits by collecting previous online shopping behavior data of users, and then guide the users to push commodity sequencing of commodities to the users in subsequent online shopping through the user portraits, so that e-commerce searches are developed from original 'one thousand of people' to today personalized 'one thousand of people'. In the traditional commodity ordering method, each time of behavior data of a user is taken as a basis for user portrait formation, and after T +1 mode calculation, commodity ordering is performed after the user searches commodities next time. However, the previous online shopping data of the user reflects the online shopping behaviors of the user at the previous online shopping time and in the online shopping environment, and although the online shopping data has a statistically significant prediction function on the favorite commodities of the user at the current time, the prediction accuracy is poor. Based on this, the commodity ordering method, the commodity ordering system and the commodity ordering device provided by the embodiment of the invention can solve the technical problem that the conventional commodity ordering method has poor prediction accuracy on the currently favored commodity of the user.
For the purpose of facilitating an understanding of the present embodiments, reference will now be made in detail to the embodiments of the present invention, examples of which are illustrated in the accompanying drawings.
An embodiment of the present invention provides a method for sorting commodities, as shown in fig. 1, including:
step S102, commodity information of commodities triggered by a user on the first commodity list page is obtained.
And step S104, after the page change instruction is received, updating the ordered list of the commodities to be displayed on the second commodity list page according to the similarity with the commodity information, wherein the second commodity list page is a commodity list page to which the first commodity list page jumps after the page change instruction.
And S106, displaying the commodities to be displayed on a second commodity list page according to the updated sorted list.
It should be noted that the commodity ordering method provided by the embodiment of the invention can be applied to a search end of an e-commerce website. The user clicks the triggered commodity on the page end of the E-commerce website, the search end obtains commodity information, and after the user sends a page changing instruction through the page end, the search end pushes the commodity to be displayed to the page end.
Specifically, the first commodity listing page and the second commodity listing page are two web pages for displaying commodities in sequence. The user can browse the first commodity list page firstly, and can click to check the commodities in the first commodity list page according to the current preference, and the commodities clicked and checked by the user are the triggered commodities. And after the user finishes browsing the first commodity list page, sending a page change instruction, and browsing the commodities on the second commodity list page.
In the commodity sorting method provided by the embodiment of the invention, after the user browses the commodities in the first commodity list page, if the page is required to be changed through the page changing instruction, the commodities which are updated and arranged according to the similarity with the commodity information are obtained on the second commodity list page, thereby enabling the shopping behavior information generated by the user on the first commodity list page to be fed back to the second commodity list page in real time, namely, the to-be-displayed commodity ordered list of the second commodity list page accurately reflects the likes and dislikes of the user at the current online shopping time and in the current online shopping environment, the purpose of feeding back the instant online shopping behavior of the user to the current commodity query transaction is achieved, the ordering of the commodities queried by the second commodity list page is optimized, the degree of engagement between the commodity ordering and the search intention of the user is improved, and the technical problem that the prediction accuracy of the current favored commodities of the users is poor by the traditional commodity ordering method is further solved.
In an optional implementation manner of the embodiment of the present invention, as shown in fig. 2, in step S104, updating the sorted list of the commodities to be displayed on the second commodity list page according to the similarity with the commodity information, where the updating includes:
step S201, extracting a sub-information set of the triggered commodity from the commodity information, wherein the sub-information set at least comprises the following sub-information: picture information, title information, price information, label information.
Specifically, the tag information includes, for example: discount, hot-sell, new, gift.
Step S202, calculating the similarity of the commodities to be displayed and each piece of sub information in the sub information set to obtain a first sub similarity, wherein the first sub similarity corresponds to the sub information one to one.
And step S203, updating the sorted list of the commodities to be displayed on the second commodity list page according to the first sub-similarity.
Specifically, the higher the similarity between the to-be-displayed commodity and each piece of sub information in the sub information set is, the larger the first sub-similarity is, and further, the higher the similarity is, the further forward the to-be-displayed commodity is displayed in the sorted list.
A typical example is: the user triggers the commodity with the label of 'money explosion is as low as 5 folds' on the first commodity list page. In this case, the sub information set extracted from the commodity information includes label information of "explosive money is as low as 5 folds"; further, according to the discount degree and the free sale degree of the commodities to be displayed, the similarity of the commodities to be displayed and sub information of 'the explosive money is as low as 5 folds' is calculated, a first sub-similarity is calculated, and the commodities to be displayed, which are high in comprehensive degree of discount and free sale, have a large first sub-similarity; next, in the second article listing page, the article to be displayed, the first sub-similarity of which is larger, is arranged at a position closer to the front.
In the embodiment of the present invention, the sub information set at least includes the following sub information: the display device comprises picture information, title information, price information and label information, so that the similarity of the commodities to be displayed and the triggered commodities is determined according to one or more of the picture information, the title information, the price information and the label information. Because the picture information, the title information, the price information and the label information have better descriptive performance on the characteristics of the triggered commodities, and most users generate interest in the commodities in view of the information in the webpage, the similarity between the commodities to be displayed and the triggered commodities is updated according to one or more of the picture information, the title information, the price information and the label information, so that the ranking list can be more accurately matched with the current online shopping time of the users and the likes and dislikes in the current online shopping environment, and the technical problem that the traditional commodity ranking method has poor precision in predicting the currently favored commodities of the users is solved.
In order to facilitate explanation of the technical scheme, it is assumed that the number of triggered commodities is 1 and is marked as Pa; and the item to be displayed of the second item list page is marked as P. The following describes the calculation of the first sub-similarity of the various sub-information of P and Pa:
(I) Picture information
In another optional implementation manner of the embodiment of the present invention, the sub information in the sub information set includes picture information, as shown in fig. 3, step S202 is to calculate similarity between the product to be displayed and each piece of sub information in the sub information set, to obtain a first sub similarity, and includes:
step S301, extracting SIFT features from the picture of the commodity to be displayed based on an SIFT algorithm, and obtaining a feature list formed by the SIFT features.
Specifically, the SIFT (Scale-invariant feature transform) algorithm is an algorithm for detecting local features, the SIFT features are local features of an image, the SIFT features keep invariance to rotation, Scale scaling and brightness change, and keep a certain degree of stability to view angle change, affine transformation and noise.
For example, a plurality of SIFT features are extracted from a picture of a commodity to be displayed, the feature list comprises the plurality of SIFT features, and the data structure of each SIFT feature can be represented by a multi-dimensional array.
The SIFT algorithm extraction of SIFT features is realized by the following steps:
(1) generating a scale space;
(2) detecting a scale space extreme point;
(3) accurately positioning an extreme point;
(4) assigning a direction parameter for each key point;
(5) and generating a key point descriptor.
And step S302, clustering the feature list by using a K-means algorithm, and determining a clustering center obtained by clustering as a standard feature of the feature list.
In particular, the K-means algorithm is a typical clustering algorithm for clustering target objects into compact and independent clusters based on distance. After the algorithm is executed, the cluster center points of each cluster are taken as the standard features of the cluster, and the dimension reduction of the target object feature list is completed. The number of the clustering centers is the dimension of the target object after dimension reduction, but the number of the clustering centers does not significantly affect the dimension reduction of the SIFT features, and the number of the clustering centers can be, for example, a square root of the sum of all SIFT feature dimensions.
The K-means algorithm is realized according to the following idea:
(1) randomly selecting k data objects from the n data objects as initial center objects of the cluster;
(2) calculating the distance between each data object and the center objects according to the mean value (center object) of each cluster, and clustering and dividing the corresponding data objects again according to the minimum distance;
(3) re-computing the mean (center object) of each cluster;
(4) and (3) circulating from (2) to (3) until the mean value of each cluster is not changed any more.
And step S303, solving the theme with the maximum standard feature occurrence probability through an L DA model, and determining the theme as the category of the commodity to be displayed.
Specifically, the L DA (L initial diagnostic Analysis) model is a typical bag-of-words model used for document classification. L DA topic model considers a document to follow a distribution and to choose multiple topics, each topic follows the same distribution and chooses multiple words, and this relationship forms a Dirichlet distribution.
The L DA model is realized by the following steps:
l DA topic model considers a document to generate words as follows:
(1) sampling a topic distribution theta i of a generated document i from a Dirichlet distribution α;
(2) sampling a topic z { i, j } of a jth word of a document i from the topic distribution theta i to generate a topic;
(3) sampling from Dirichlet distribution β to generate a word distribution φ z { i, j } of subject z { i, j };
(4) sampling from the word distribution phi z i, j ultimately generates words w i, j.
The process of solving the probability of the occurrence of the largest topic using the Gibbs Sampling method is as follows:
first, the following four steps are performed:
(1) traversing all words in all documents, and randomly distributing a theme for each word;
(2) traversing all words in the document, taking out the current word for each traversed word, then sampling a new theme according to the probability distribution of a topic sample in L DA, and updating the number of themes in the document, the occurrence frequency of each theme, the number of words and the occurrence frequency of each word;
(3) the topoic sample probability distribution is updated accordingly;
(4) and (3) circulating the steps (2) and (3) until the theme-word parameter matrix phi and the document-theme matrix theta are not changed any more.
Then, based on the document-topic matrix obtained by solving, the topic with the highest probability of occurrence under the document is selected as the category of the document.
For the picture to be displayed in the embodiment of the invention, the theme with the largest occurrence probability is the category of the picture to be displayed.
And step S304, determining a first sub-similarity corresponding to the commodity to be displayed and the picture information according to the category of the commodity to be displayed and the similarity of the picture information.
Specifically, a first sub-similarity related to the picture information may be denoted as Sp, and if the category of the to-be-displayed product is similar to the picture information, Sp may be denoted as 1; otherwise, let Sp equal to 0.
In the embodiment of the invention, the SIFT algorithm, the K-means algorithm and the L DA model are combined, so that the calculation of the first sub-similarity between the to-be-displayed commodity and the picture information is accurately realized.
It should be emphasized that the SIFT algorithm is an optional local feature detection algorithm for obtaining the feature list in the present invention, the K-means algorithm is an optional clustering algorithm for obtaining the standard feature in the present invention, and the L DA model is an optional bag-of-words model for obtaining the category of the commodity to be displayed in the present invention.
(II) header information
The title information is used as text, and the L DA algorithm mentioned above can be used to calculate the first sub-similarity between the item to be displayed and the title information.
(III) price information
The price information is digital information, the span range is large, and the K-means algorithm can be adopted for calculating the first sub-similarity of the commodity to be displayed and the price information.
Specifically, the K-means algorithm is adopted to calculate the first sub-similarity of the commodities to be displayed and price information, single-dimensional clustering taking price as a characteristic is essential, the number of the clustering centers is determined according to the type of the inquired commodities, and then the number of the clustering centers is manually configured according to the type of the inquired commodities. The first sub-similarity related to the price information may be denoted as Spri, and if the category of the product to be displayed and the price information belong to the same cluster, Spri may be made 1; otherwise, let Spri equal to 0.
(IV) tag information
The following formula can be adopted for calculating the first sub-similarity between the to-be-displayed goods and the label information:
for example, if the Pa label is discount, hot-sell, new product, and the P label is discount, hot-sell, the intersection between the Pa label and the P label is discount, hot-sell, and therefore the number of intersections between the Pa label and the P label is 2, and Stg is 2/3.
In the case that the number of the sub information in the sub information set is multiple, another optional implementation manner of the embodiment of the present invention provides an implementation manner in which, in step S203, the sorted list of the commodities to be displayed on the second commodity list page is updated according to the first sub similarity. As shown in fig. 4, the method includes:
step S401, obtaining a first sub-similarity between a current commodity to be displayed and a plurality of pieces of sub-information, and obtaining a plurality of first sub-similarities of the current commodity to be displayed;
and S402, updating the arrangement sequence of the current commodities to be displayed in the ordered list according to the plurality of first sub-similarities to obtain an updated ordered list.
Specifically, the arrangement order of the current commodities to be displayed in the ordered list can be updated according to the sum of the first sub-similarities, so that the commodities to be displayed with the larger sum of the first sub-similarities are displayed at the front position of the second commodity list page.
The number of sub information is exemplarily illustrated as 2. The two pieces of sub information may be any two different pieces of information among picture information, title information, price information, and tag information, and if the two pieces of sub information are picture information and title information, the plurality of first sub similarities are: and if the first sub-similarity between the current to-be-displayed commodity and the picture information is the first sub-similarity between the current to-be-displayed commodity and the title information, updating the arrangement sequence of the current to-be-displayed commodity in the ordered list according to the two first sub-similarities in step S402.
According to the embodiment of the invention, the arrangement sequence of the current commodities to be displayed in the ordered list is updated according to the plurality of first sub-similarities, so that the ordered list more comprehensively conforms to the current online shopping time of the user and the likes and dislikes in the current online shopping environment, and the technical problem that the conventional commodity ordering method has poor prediction accuracy on the current favored commodities of the user is solved.
In another optional implementation manner of the embodiment of the present invention, in step S402, updating an arrangement order of the current commodities to be displayed in the ordered list according to the plurality of first sub-similarities to obtain an updated ordered list, where the updating includes:
acquiring weights corresponding to the sub-information to obtain a plurality of sub-weights, wherein the sub-weights correspond to the sub-information one to one;
calculating a weighted sum of a plurality of first sub-similarities based on a product of the corresponding sub-weight of each piece of sub-information and the first sub-similarity;
and updating the arrangement sequence of the current commodities to be displayed in the ordered list according to the weighted sum.
Specifically, the weighted sum larger article to be displayed may be displayed at the front position of the second article listing page.
If the picture information corresponding sub-weight Wp and the first sub-similarity Sp, the title information corresponding sub-weight Wt and the first sub-similarity St, the price information corresponding sub-weight Wpri and the first sub-similarity Spri, and the label information corresponding sub-weight Wtag and the first sub-similarity Stag are taken as the sub-information, the sub-information includes: in the case of picture information, title information, price information, and label information, the weighted sum is: score ═ Sp × Wp + St × Wt + Spri × Wpri + Stag × Wtag.
It should be noted that the weights corresponding to the sub information may be determined according to the degree of emphasis of the user on the sub information, and the more emphasized the user is, the greater the corresponding weight is. Since most of the users in online shopping have the types of commodities to be purchased, and the picture information in the commodity information mostly represents the types of the commodities, a larger numerical value can be configured for the corresponding sub-weight Wp of the picture information.
In the embodiment of the invention, the arrangement sequence of the current commodities to be displayed in the ordered list is updated according to the weighted sum, so that the arrangement sequence of the current commodities to be displayed is determined by the corresponding first sub-similarity of each sub-information according to different proportions, the arrangement of the sub-weights is favorable for considering the attention degree of a user to different sub-information, the arrangement is more scientific and reasonable, and the arrangement sequence of the current commodities to be displayed can better accord with the current online shopping time of the user and the likes and dislikes degree in the current online shopping environment.
In a case that the number of the triggered products Pa is multiple, another optional implementation manner of the embodiment of the present invention provides an implementation manner in which, in step S104, the sorted list of the products to be displayed on the second product list page is updated according to the similarity with the product information, and as shown in fig. 5 in detail, the implementation manner includes:
step S501, obtaining the similarity between the current commodity to be displayed and a plurality of triggered commodities, and obtaining a plurality of second sub-similarities of the current commodity to be displayed;
and step S502, updating the arrangement sequence of the current commodities to be displayed in the ordered list according to the sum of the second sub-similarities to obtain an updated ordered list.
Specifically, assume that the number of triggered commodities is N, that is, the method includes: pa1, Pa2 … … Pan, wherein the second sub-similarity between the current commodity to be displayed and the Pai is recorded as
The sum of the second sub-similarities is:
thus, step S502 determines the arrangement order of the current goods to be displayed in the ordered list according to TotalScore. For example, if the sum of the second sub-similarities of the product P1 to be displayed is TotalScore (P1) equal to 7, and the sum of the second sub-similarities of the product Pi to be currently displayed is TotalScore (Pi) equal to 3, the arrangement order of the product P1 to be currently displayed is located after the product P to be displayed.
In the embodiment of the invention, the number of the triggered commodities Pa is multiple, and the multiple triggered commodities Pa reflect the current online shopping time of the user and the search intention in the current online shopping environment for multiple times, so that the sum of the multiple second sub-similarities can more accurately determine the arrangement sequence of the current commodities to be displayed, and the arrangement sequence of the current commodities to be displayed can more accord with the current online shopping time of the user and the preference degree in the current online shopping environment.
An embodiment of the present invention provides a commodity ordering system, as shown in fig. 6, including:
the acquisition module 100 is configured to acquire commodity information of a commodity triggered by a user on a first commodity list page;
the updating module 200 is configured to update the ordered list of the commodities to be displayed on the second commodity list page according to the similarity between the received page change instruction and the commodity information, where the second commodity list page is a commodity list page to which the first commodity list page jumps after the page change instruction;
the display module 300 is configured to display the to-be-displayed commodities on the second commodity list page according to the updated sorted list.
According to the commodity ordering system provided by the embodiment of the invention, after a user browses commodities on the first commodity list page, if the page is to be changed through a page changing instruction, commodities arranged according to the similarity with the commodity information are obtained on the second commodity list page, so that the shopping behavior information generated on the first commodity list page by the user is fed back to the second commodity list page in real time, the purpose that the instant online shopping behavior of the user is fed back to the current commodity query transaction is achieved, the ordering of the queried commodities is optimized, the degree of conformity between the commodity ordering and the search intention of the user is improved, and the technical problem that the accuracy of the traditional commodity ordering method for predicting currently favored commodities of the user is poor is further solved.
In an optional implementation manner of the embodiment of the present invention, the update module includes:
the extraction unit is used for extracting a sub-information set of the triggered commodity from the commodity information, wherein the sub-information set at least comprises the following sub-information: picture information, title information, price information, label information;
the calculating unit is used for calculating the similarity of the commodities to be displayed and each piece of sub information in the sub information set to obtain a first sub similarity, wherein the first sub similarity corresponds to the sub information one by one;
and the updating unit is used for updating the sorted list of the commodities to be displayed on the second commodity list page according to the first sub-similarity.
In another optional implementation manner of the embodiment of the present invention, the number of pieces of sub information in the sub information set is multiple, and the updating unit includes:
the acquiring subunit is used for acquiring a first sub-similarity between the current to-be-displayed commodity and the plurality of pieces of sub-information to obtain a plurality of first sub-similarities of the current to-be-displayed commodity;
and the updating subunit is used for updating the arrangement sequence of the current commodities to be displayed in the ordered list according to the plurality of first sub-similarities so as to obtain the ordered list.
In another optional implementation manner of the embodiment of the present invention, the update subunit is configured to:
acquiring weights corresponding to the sub-information to obtain a plurality of sub-weights, wherein the sub-weights correspond to the sub-information one to one;
calculating a weighted sum of a plurality of first sub-similarities based on a product of the corresponding sub-weight of each piece of sub-information and the first sub-similarity;
and updating the arrangement sequence of the current commodities to be displayed in the ordered list according to the weighted sum.
In another optional implementation manner of the embodiment of the present invention, the sub information in the sub information set includes picture information, and the calculating unit is configured to:
extracting local features from a picture of a commodity to be displayed based on a local feature detection algorithm to obtain a feature list formed by the local features;
clustering the feature list by using a clustering algorithm, and determining a clustering center obtained by clustering as a standard feature of the feature list;
solving the theme with the maximum standard feature occurrence probability through a bag-of-words model, and determining the theme as the category of the commodity to be displayed;
and determining the corresponding first sub-similarity of the commodity to be displayed and the picture information according to the category of the commodity to be displayed and the similarity of the picture information.
In another alternative implementation of an embodiment of the present invention,
the local feature detection algorithm adopts an SIFT algorithm; and/or the presence of a gas in the gas,
the clustering algorithm adopts a K-means algorithm; and/or the presence of a gas in the gas,
the bag of words model uses the L DA model.
In another optional implementation manner of the embodiment of the present invention, the number of triggered commodities is multiple, and the update module is configured to:
acquiring the similarity between the current to-be-displayed commodity and the plurality of triggered commodities to obtain a plurality of second sub-similarities of the current to-be-displayed commodity;
and updating the arrangement sequence of the current commodities to be displayed in the ordered list according to the sum of the second sub-similarities to obtain an updated ordered list.
The embodiment of the invention provides a computer-readable storage medium, which stores computer instructions, and when the computer instructions are executed, the commodity ordering method of the first embodiment is realized.
Specifically, the readable storage medium includes: a U-disk, a removable hard disk, a Read-only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
When the computer instruction provided by the embodiment of the present invention is executed, the method for sorting commodities in the first embodiment of the present invention is implemented, specifically, commodity information of a commodity triggered by a user on a first commodity list page is obtained; after a page change instruction is received, determining a ranked list of commodities to be displayed on a second commodity list page according to the similarity of the commodity information and the page change instruction, wherein the second commodity list page is a commodity list page to which the first commodity list page jumps after the page change instruction; and displaying the commodities to be displayed on a second commodity list page according to the updated sorted list. Based on the commodity ordering method, after a user browses commodities on a first commodity list page, if the page is to be changed through a page changing instruction, commodities arranged according to the similarity with the commodity information are obtained on a second commodity list page, so that the shopping behavior information generated on the first commodity list page by the user is fed back to the second commodity list page in real time, the purpose that the instant online shopping behavior of the user is fed back to the current commodity query transaction is achieved, the ordering of the commodities queried on the second commodity list page is optimized, the degree of conformity between the commodity ordering and the search intention of the user is improved, and the technical problem that the current favored commodity of the user is predicted to be poor by a traditional commodity ordering method is solved.
An apparatus for sorting commodities provided in an embodiment of the present invention, as shown in fig. 7, includes:
a memory 701 for storing computer instructions;
a processor 702 coupled to the memory 701, wherein the processor 702 is configured to execute the method for ordering items according to the first embodiment based on the computer instructions stored in the memory 701.
Specifically, the memory 701 and the processor 702 may be connected to the input/output device 703 through a bus, the memory 701 may store various computer instructions and data required for performing system functions, and the processor 702 may read various computer instructions from the memory 701 to perform various appropriate actions and processes.
According to the commodity sequencing device provided by the embodiment of the invention, after a user browses commodities on the first commodity list page, if the page is to be changed through a page changing instruction, commodities arranged according to the similarity with the commodity information are obtained on the second commodity list page, so that the shopping behavior information generated on the first commodity list page by the user is fed back to the second commodity list page in real time, the purpose that the instant online shopping behavior of the user is fed back to the current commodity query transaction is achieved, the sequencing of the queried commodities is optimized, the degree of conformity between the commodity sequencing and the search intention of the user is improved, and the technical problem that the current favored commodity of the user is predicted with poor precision by a traditional commodity sequencing method is solved.
The flowcharts and block diagrams in the figures and block diagrams illustrate the possible architectures, functions, and operations of the systems, methods, and apparatuses according to the embodiments of the present invention, and may represent a module, a program segment, or merely a code segment, which is an executable instruction for implementing a specified logical function. It should also be noted that the executable instructions that implement the specified logical functions may be recombined to create new modules and program segments. The blocks of the drawings, and the order of the blocks, are thus provided to better illustrate the processes and steps of the embodiments and should not be taken as limiting the invention itself.
Although the steps and sequence of steps of the embodiments of the present invention are presented in method and method illustrations, the steps implementing the specified logical functions may be re-combined to create new steps.
Systems and methods according to the present invention may be deployed on a single server or on multiple servers. For example, different modules may be deployed on different servers, respectively, to form a dedicated server. Alternatively, the same functional unit, module or system may be deployed in a distributed fashion across multiple servers to relieve load stress. The server includes but is not limited to a plurality of PCs, PC servers, blades, supercomputers, etc. on the same local area network and connected via the Internet. It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the system and the apparatus described above may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.
Finally, it should be noted that: the above-mentioned embodiments are only specific embodiments of the present invention, which are used for illustrating the technical solutions of the present invention and not for limiting the same, and the protection scope of the present invention is not limited thereto, although the present invention is described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: any person skilled in the art can modify or easily conceive the technical solutions described in the foregoing embodiments or equivalent substitutes for some technical features within the technical scope of the present disclosure; such modifications, changes or substitutions do not depart from the spirit and scope of the embodiments of the present invention, and they should be construed as being included therein. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (16)
1. A method of ordering articles, comprising:
acquiring commodity information of a commodity triggered by a user on a first commodity list page;
after a page change instruction is received, updating a ranked list of commodities to be displayed on a second commodity list page according to the similarity of the commodity information and the page change instruction, wherein the second commodity list page is a commodity list page to which the first commodity list page jumps after the page change instruction;
and displaying the commodities to be displayed on the second commodity list page according to the updated sorted list.
2. The method of claim 1, wherein updating the ordered list of items to be displayed on the second item list page according to the similarity to the item information comprises:
extracting a sub-information set of the triggered commodity from the commodity information, wherein the sub-information set at least comprises the following sub-information: picture information, title information, price information, label information;
calculating the similarity of the commodities to be displayed and each piece of sub information in the sub information set to obtain a first sub similarity, wherein the first sub similarity corresponds to the sub information one to one;
and updating the sorted list of the commodities to be displayed on the second commodity list page according to the first sub-similarity.
3. The method of claim 2, wherein the number of the sub information in the sub information set is plural, and updating the ordered list of the commodities to be displayed on the second commodity list page according to the first sub similarity comprises:
acquiring the first sub-similarity between the current to-be-displayed commodity and the plurality of pieces of sub-information to obtain a plurality of first sub-similarities of the current to-be-displayed commodity;
and updating the arrangement sequence of the current commodities to be displayed in the ordered list according to the plurality of first sub-similarities to obtain the updated ordered list.
4. The method according to claim 3, wherein updating the arrangement order of the current to-be-displayed commodities in the ordered list according to the plurality of first sub-similarities to obtain the updated ordered list comprises:
acquiring weights corresponding to the sub information to obtain a plurality of sub weights, wherein the sub weights correspond to the sub information one to one;
calculating a weighted sum of a plurality of first sub-similarities based on the product of the sub-weight corresponding to each piece of the sub-information and the first sub-similarity;
and updating the arrangement sequence of the current commodities to be displayed in the ordered list according to the weighted sum.
5. The method according to claim 2, wherein the sub information in the sub information set includes picture information, and the calculating of the similarity between the to-be-displayed commodity and each sub information in the sub information set to obtain a first sub similarity includes:
extracting local features from the picture of the commodity to be displayed based on a local feature detection algorithm to obtain a feature list formed by the local features;
clustering the feature list by using a clustering algorithm, and determining a clustering center obtained by clustering as a standard feature of the feature list;
obtaining a theme with the maximum standard feature occurrence probability through a bag-of-words model, and determining the theme as the category of the commodity to be displayed;
and determining a first sub-similarity corresponding to the commodity to be displayed and the picture information according to the category of the commodity to be displayed and the similarity of the picture information.
6. The method of claim 5,
the local feature detection algorithm adopts an SIFT algorithm; and/or the presence of a gas in the gas,
the clustering algorithm adopts a K-means algorithm; and/or the presence of a gas in the gas,
the bag of words model adopts an L DA model.
7. The method of claim 1, wherein the number of the triggered commodities is multiple, and the updating of the ordered list of commodities to be displayed on the second commodity list page according to the similarity with the commodity information comprises:
obtaining the similarity between the current to-be-displayed commodity and the plurality of triggered commodities to obtain a plurality of second sub-similarities of the current to-be-displayed commodity;
and updating the arrangement sequence of the current commodities to be displayed in the ordered list according to the sum of the second sub-similarities to obtain the updated ordered list.
8. A merchandise sequencing system, comprising:
the acquisition module is used for acquiring commodity information of a commodity triggered by a user on a first commodity list page;
the updating module is used for updating the ordered list of the commodities to be displayed on a second commodity list page according to the similarity with the commodity information after receiving a page change instruction, wherein the second commodity list page is a commodity list page to which the first commodity list page jumps after the page change instruction;
and the display module is used for displaying the commodities to be displayed on the second commodity list page according to the updated sorted list.
9. The system of claim 8, wherein the update module comprises:
an extracting unit, configured to extract a sub-information set of the triggered product from the product information, where the sub-information set at least includes one of the following pieces of sub-information: picture information, title information, price information, label information;
the calculation unit is used for calculating the similarity of the commodities to be displayed and each piece of sub information in the sub information set to obtain a first sub similarity, wherein the first sub similarity is in one-to-one correspondence with the sub information;
and the updating unit is used for updating the sorted list of the commodities to be displayed on the second commodity list page according to the first sub-similarity.
10. The system according to claim 9, wherein the number of the sub information in the sub information set is plural, and the updating unit comprises:
the obtaining subunit is configured to obtain the first sub-similarity between the current to-be-displayed commodity and the plurality of pieces of sub-information, so as to obtain a plurality of first sub-similarities of the current to-be-displayed commodity;
and the updating subunit is configured to update the arrangement order of the current commodities to be displayed in the ordered list according to the plurality of first sub-similarities, so as to obtain the updated ordered list.
11. The system of claim 10, wherein the update subunit is configured to:
acquiring weights corresponding to the sub information to obtain a plurality of sub weights, wherein the sub weights correspond to the sub information one to one;
calculating a weighted sum of a plurality of first sub-similarities based on the product of the sub-weight corresponding to each piece of the sub-information and the first sub-similarity;
and updating the arrangement sequence of the current commodities to be displayed in the ordered list according to the weighted sum.
12. The system of claim 9, wherein the sub-information in the sub-information set comprises picture information, and wherein the computing unit is configured to:
extracting local features from the picture of the commodity to be displayed based on a local feature detection algorithm to obtain a feature list formed by the local features;
clustering the feature list by using a clustering algorithm, and determining a clustering center obtained by clustering as a standard feature of the feature list;
obtaining a theme with the maximum standard feature occurrence probability through a bag-of-words model, and determining the theme as the category of the commodity to be displayed;
and determining a first sub-similarity corresponding to the commodity to be displayed and the picture information according to the category of the commodity to be displayed and the similarity of the picture information.
13. The system of claim 12,
the local feature detection algorithm adopts an SIFT algorithm; and/or the presence of a gas in the gas,
the clustering algorithm adopts a K-means algorithm; and/or the presence of a gas in the gas,
the bag of words model adopts an L DA model.
14. The system of claim 8, wherein the number of triggered items is multiple, and the update module is configured to:
obtaining the similarity between the current to-be-displayed commodity and the plurality of triggered commodities to obtain a plurality of second sub-similarities of the current to-be-displayed commodity;
and updating the arrangement sequence of the current commodities to be displayed in the ordered list according to the sum of the second sub-similarities to obtain the updated ordered list.
15. A computer-readable storage medium, characterized in that the computer-readable storage medium stores computer instructions which, when executed, implement the method of ordering articles according to any of claims 1 to 7.
16. An article sequencing device, comprising:
a memory for storing computer instructions;
a processor coupled to the memory, the processor configured to perform a method of implementing the item ordering method of any one of claims 1-7 based on computer instructions stored by the memory.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910045121.0A CN111445302A (en) | 2019-01-17 | 2019-01-17 | Commodity sorting method, system and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910045121.0A CN111445302A (en) | 2019-01-17 | 2019-01-17 | Commodity sorting method, system and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111445302A true CN111445302A (en) | 2020-07-24 |
Family
ID=71648810
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910045121.0A Pending CN111445302A (en) | 2019-01-17 | 2019-01-17 | Commodity sorting method, system and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111445302A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112116421A (en) * | 2020-09-09 | 2020-12-22 | 江苏亿讯网络科技有限公司 | Method and system for optimizing and recommending commodities in self-built welfare mall |
| CN113744015A (en) * | 2020-10-20 | 2021-12-03 | 北京沃东天骏信息技术有限公司 | Sorting method, device, equipment and computer storage medium |
-
2019
- 2019-01-17 CN CN201910045121.0A patent/CN111445302A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112116421A (en) * | 2020-09-09 | 2020-12-22 | 江苏亿讯网络科技有限公司 | Method and system for optimizing and recommending commodities in self-built welfare mall |
| CN113744015A (en) * | 2020-10-20 | 2021-12-03 | 北京沃东天骏信息技术有限公司 | Sorting method, device, equipment and computer storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3143523B1 (en) | Visual interactive search | |
| US10410224B1 (en) | Determining item feature information from user content | |
| US10204121B1 (en) | System and method for providing query recommendations based on search activity of a user base | |
| CN1694105B (en) | Method and system for identifying image relatedness using link and page layout analysis | |
| CN101248435B (en) | Determination of a desired repository | |
| JP5962926B2 (en) | Recommender system, recommendation method, and program | |
| TWI567673B (en) | Sorting methods and devices for search results | |
| US20110125791A1 (en) | Query classification using search result tag ratios | |
| WO2008106667A1 (en) | Searching heterogeneous interrelated entities | |
| US20100106719A1 (en) | Context-sensitive search | |
| EP3238099B1 (en) | Method and user interface for presenting auxiliary content together with image search results | |
| US20080294643A1 (en) | Method of generating information related to a web page | |
| JP2019125007A (en) | Information analyzer, information analysis method and information analysis program | |
| CN101246484A (en) | Electric text similarity processing method and system convenient for query | |
| US20140081701A1 (en) | Determining and using brand information in electronic commerce | |
| CN108664637A (en) | A kind of search method and system | |
| US8463770B1 (en) | System and method for conditioning search results | |
| CN114443847A (en) | Text classification method, text processing method, text classification device, text processing device, computer equipment and storage medium | |
| CN111445302A (en) | Commodity sorting method, system and device | |
| CN111915414B (en) | Method and device for displaying target object sequence to target user | |
| TWI507902B (en) | A retrieval system, a retrieval condition setting device, a control method of a search condition setting device, a computer program product, and an information memory medium | |
| JP4667889B2 (en) | Data map creation server and data map creation program | |
| JP6433270B2 (en) | Content search result providing system and content search result providing method | |
| An et al. | Neural ctr prediction for native ad | |
| KR20230014035A (en) | Method and device for recommending related documents through user search intent analysis |
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 |