CN113592589A - Textile raw material recommendation method and device and processor - Google Patents
Textile raw material recommendation method and device and processor Download PDFInfo
- Publication number
- CN113592589A CN113592589A CN202110850576.7A CN202110850576A CN113592589A CN 113592589 A CN113592589 A CN 113592589A CN 202110850576 A CN202110850576 A CN 202110850576A CN 113592589 A CN113592589 A CN 113592589A
- Authority
- CN
- China
- Prior art keywords
- raw material
- user
- commodity
- determining
- commodities
- 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.)
- Granted
Links
- 239000002994 raw material Substances 0.000 title claims abstract description 245
- 238000000034 method Methods 0.000 title claims abstract description 51
- 239000004753 textile Substances 0.000 title claims abstract description 47
- 239000000463 material Substances 0.000 claims abstract description 34
- 230000006399 behavior Effects 0.000 claims description 72
- 239000013598 vector Substances 0.000 claims description 9
- 230000003542 behavioural effect Effects 0.000 claims description 5
- 238000000605 extraction Methods 0.000 claims description 5
- 238000003860 storage Methods 0.000 abstract description 16
- 238000010586 diagram Methods 0.000 description 13
- 238000004590 computer program Methods 0.000 description 11
- 230000006870 function Effects 0.000 description 11
- 230000008569 process Effects 0.000 description 9
- 238000004364 calculation method Methods 0.000 description 7
- 238000012545 processing Methods 0.000 description 7
- 238000004422 calculation algorithm Methods 0.000 description 4
- 238000001914 filtration Methods 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 229920000742 Cotton Polymers 0.000 description 2
- 230000009471 action Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000004744 fabric Substances 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 230000001680 brushing effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000005315 distribution function Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000008707 rearrangement Effects 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
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]
- G06Q30/0631—Item recommendations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- 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
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
-
- 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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/04—Manufacturing
-
- 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
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Accounting & Taxation (AREA)
- Strategic Management (AREA)
- Finance (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Development Economics (AREA)
- General Business, Economics & Management (AREA)
- Economics (AREA)
- Marketing (AREA)
- Data Mining & Analysis (AREA)
- Entrepreneurship & Innovation (AREA)
- Evolutionary Biology (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Manufacturing & Machinery (AREA)
- Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Primary Health Care (AREA)
- Tourism & Hospitality (AREA)
- Life Sciences & Earth Sciences (AREA)
- Game Theory and Decision Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the application provides a textile material recommending method, a textile material recommending device, a processor and a storage medium. The method comprises the following steps: acquiring behavior data of a user; determining the interest degree of the users in each dimension aiming at the textile raw material commodities and the user similarity among the users according to the behavior data; determining a first recommendation candidate set for the user according to the interest degree; determining the raw material similarity between raw material commodities; determining a second recommended candidate set of the user according to the user similarity and the raw material similarity; inputting at least one of the behavior data of the user, the raw material characteristics of each textile raw material and the context information into a prediction model to determine a third recommended candidate set of the user; determining a final recommendation set of the user according to the first recommendation candidate set, the second recommendation candidate set and the third recommendation candidate set; and determining the raw material commodities included in the final recommendation set as recommended raw material commodities for the user.
Description
Technical Field
The application relates to the technical field of computers, in particular to a textile raw material recommending method, a textile raw material recommending device, a storage medium and a processor.
Background
The company has a raw material supply service and a gray cloth production platform service, and hopes to optimize raw material recommendation sequencing for a user through factory data, raw material related data, user behavior of a raw material recommendation position and the like of gray cloth production so as to improve the raw material click rate and further improve the transaction rate.
The existing industry is simple in implementation mode, and generally, relevant dimension data is solved through ETL (extract transform and load) processing, after normalization, weighting and summing are performed to solve a comprehensive score, and then reverse order output is performed. The method for calculating the score in the fracture area to recommend the raw materials has low recommendation accuracy and can not realize intelligent recommendation.
Disclosure of Invention
The embodiment of the application aims to provide a textile material recommending method, a textile material recommending device, a storage medium and a processor.
In order to achieve the above object, a first aspect of the present application provides a textile material recommendation method, including:
acquiring behavior data of a user;
determining the interest degree of the users in each dimension aiming at the textile raw material commodities and the user similarity among the users according to the behavior data;
determining a first recommendation candidate set for the user according to the interest degree;
determining the raw material similarity between raw material commodities;
Determining a second recommended candidate set of the user according to the user similarity and the raw material similarity;
inputting at least one of the behavior data of the user, the raw material characteristics of each textile raw material and the context information into a prediction model to determine a third recommended candidate set of the user;
determining a final recommendation set of the user according to the first recommendation candidate set, the second recommendation candidate set and the third recommendation candidate set;
and determining the raw material commodities included in the final recommendation set as recommended raw material commodities for the user.
Optionally, the behavior data includes operation behaviors of the user for the raw material commodity in a plurality of time periods of different durations; determining the interest degree of the users in the textile raw material commodities and the user similarity among the users according to the behavior data comprises the following steps: acquiring commodity attributes of raw material commodities corresponding to each operation behavior of a user; and determining the interest degree of the user for each raw material commodity according to the operation behaviors and the commodity attributes.
Optionally, the item attributes include item keywords; determining the interest degree of the user for each raw material commodity according to the operation behaviors and the commodity attributes comprises the following steps: determining interest scores of the users for the commodity keywords of the raw material commodities aiming at each raw material commodity; ranking the interest scores to determine a preference sequence of the user for the commodity keywords; and determining the interest degree of the user in the keyword dimension for each raw material commodity according to the preference sequence.
Optionally, determining the interest score of the user for the item keyword of the raw item comprises: the interest score is determined according to the following formula:
wherein, scorekThe interest score of the user on the keyword k is indicated, i refers to the raw material commodity operated by the user i in the behavior data, aiMeans the action weight, w, corresponding to the raw material commodity of the ith operationiIs the weight of the keyword k to the material commodity of the ith operation, f (t)i) The time decay function shows the interest decay degree of the user on the raw material commodity of the ith operation at the time t.
Optionally, determining the raw material similarity between the raw material commodities comprises: acquiring raw material data of each raw material commodity; determining a characteristic vector of each raw material commodity according to the raw material data; and clustering the feature vectors to determine the raw material similarity between the raw material commodities.
Optionally, determining the material commodities included in the final recommendation set as recommended material commodities for the user further includes: sorting the raw material commodities included in the final recommendation set according to the sequence of the recommendation degrees from high to low; adjusting the sequence and recommended quantity of raw material commodities belonging to the same type; selecting a preset number of raw material commodities with the serial numbers in front to determine the raw material commodities as recommended raw material commodities of the user.
Optionally, the method further comprises: performing feature extraction on the behavior data to determine corresponding user features; determining the commodity characteristics of the raw material commodities corresponding to each operation behavior contained in the acquired behavior data; determining the characteristic weight of each user characteristic according to the commodity characteristics and the behavior data; and adjusting the recommended raw material commodity aiming at the user through the characteristic weight.
Optionally, the method further comprises: acquiring a real-time operation log of a user; determining at least one of recommended exposure rate and exposure click rate of the raw material commodity according to the operation log; and adjusting the recommended raw material commodity aiming at the user according to the recommended exposure rate and the exposure click rate.
A second aspect of the present application provides a processor configured to perform the above-described textile material recommendation method.
A third aspect of the application provides a textile material recommendation device comprising a processor as described above.
A fourth aspect of the present application provides a machine-readable storage medium having instructions stored thereon, which when executed by a processor, cause the processor to be configured to perform the above-described textile material recommendation method.
According to the method for recommending the textile raw materials, the factory capacity big data of the textile industry is utilized, the requirements of individual users and authenticated enterprise users on more accurate raw material matching recommendation are met according to the matching relations of user images, raw material commodity images and the like, and more intelligent raw material commodity recommendation is achieved.
Additional features and advantages of embodiments of the present application will be described in detail in the detailed description which follows.
Drawings
The accompanying drawings, which are included to provide a further understanding of the embodiments of the disclosure and are incorporated in and constitute a part of this specification, illustrate embodiments of the disclosure and together with the description serve to explain the embodiments of the disclosure, but are not intended to limit the embodiments of the disclosure. In the drawings:
FIG. 1 schematically illustrates a flow diagram of a method of textile material recommendation in accordance with an embodiment of the present application;
FIG. 2 schematically illustrates a schematic diagram of a textile material data recall in accordance with an embodiment of the present application;
fig. 3 schematically shows an internal structure diagram of a computer device according to an embodiment of the present application.
Detailed Description
The following detailed description of embodiments of the present application will be made with reference to the accompanying drawings. It should be understood that the detailed description and specific examples, while indicating the embodiments of the application, are given by way of illustration and explanation only, not limitation.
Fig. 1 schematically shows a flow diagram of a method for recommending textile raw materials according to an embodiment of the present application. As shown in fig. 1, in an embodiment of the present application, there is provided a method for recommending textile materials, including the steps of:
And 102, determining the interest degree of the users in various dimensions for the textile raw material commodities and the user similarity among the users according to the behavior data.
Firstly, the behavior data of the user can be acquired through the log full record of the system. The behavior data of the user comprises operation behaviors of the user for the raw material commodity in a plurality of time periods with different time lengths. Effective behaviors in the operation behaviors comprise operations of browsing, clicking, playing, commenting, forwarding and the like. By acquiring the behavior data of the user within a period of time, the interest degree of the user in each dimension for the textile raw material commodity can be determined, and the user similarity among different users can also be determined. Therefore, the preference of the user to a specific dimension can be calculated in a statistical mining mode, the interest preferences of different users are carved in a weighted list mode, the first step of thousands of people personalized recommendation is achieved, data are visual, and interpretability is strong. The obtained user portrait data can be used for other services, and a foundation is laid for personalized recommendation of other services.
In one embodiment, determining the user interest level in the textile good and the user similarity between the users from the behavior data comprises: acquiring commodity attributes of raw material commodities corresponding to each operation behavior of a user; and determining the interest degree of the user for each raw material commodity according to the operation behaviors and the commodity attributes.
Specifically, after behavior data of the user is acquired, a raw material commodity corresponding to each operation behavior of the user and a commodity attribute corresponding to the raw material commodity can be acquired. And then determining the interest degree of the user for each raw material commodity according to the operation behaviors and the commodity attributes. Further, the commodity attribute includes a commodity keyword. Determining the interest degree of the user for each raw material commodity according to the operation behaviors and the commodity attributes comprises the following steps: determining interest scores of the users for the commodity keywords of the raw material commodities aiming at each raw material commodity; ranking the interest scores to determine a preference sequence of the user for the commodity keywords; and determining the interest degree of the user in the keyword dimension for each raw material commodity according to the preference sequence. Specifically, the click operation of the user and the raw material commodity corresponding to the click operation within a period of time may be counted. The interest degree of the user on a certain attribute value can be calculated through the attribute weight of the raw material commodity. For deep behaviors such as praise, comment and collection, the user can be shown to have better preference for the contents, and weighting processing can be carried out for the operations. The interest score may be determined according to the following formula:
Wherein, scorekThe interest score of the user on the keyword k is indicated, i refers to the raw material commodity operated by the user i in the behavior data, aiFor example, the weight of the approval operation may be set to be higher than the weight of the click operation. w is aiIs the weight of the keyword k to the material commodity of the ith operation, f (t)i) The time attenuation function shows the interest attenuation degree of the raw material commodity of the ith operation at the time t, and the longer the current time is, the higher the interest attenuation degree is. The denominator is the number of raw materials for all behaviors. All keywords are normalized and sorted after score values are calculatedAnd the preference sequence of the user on the interest of the keyword can be obtained. The other dimension-dependent calculation modes are similar, but each dimension calculation result is a weighted multi-value list.
After the interest scores of the user for the commodity keywords of each raw material commodity are calculated, the interest scores can be ranked, so that a preference sequence of the user for the commodity keywords can be determined, and then the interest degree of the user for each raw material commodity in the keyword dimension can be determined according to the preference sequence.
And 104, determining the raw material similarity among the raw material commodities.
And 105, determining a second recommended candidate set of the user according to the user similarity and the raw material similarity.
And 106, inputting at least one of the behavior data of the user, the raw material characteristics of each textile raw material and the context information into a prediction model to determine a third recommended candidate set of the user.
And step 107, determining a final recommendation set of the user according to the first recommendation candidate set, the second recommendation candidate set and the third recommendation candidate set.
And step 108, determining the raw material commodities included in the final recommendation set as recommended raw material commodities for the user.
After the interest level of the user for each raw material commodity in the keyword dimension is determined, a first recommended candidate set for the user, that is, a recommended candidate set for the raw material commodity of the user, may be determined according to the interest level of the user. The determination of raw material similarity between each raw material commodity may then proceed. Further, in one embodiment, determining raw material similarity between raw material commodities comprises: acquiring raw material data of each raw material commodity; determining a characteristic vector of each raw material commodity according to the raw material data; and clustering the feature vectors to determine the raw material similarity between the raw material commodities.
Specifically, as shown in fig. 2, a schematic flow chart of data recall is given. In this embodiment, the similarity between the materials (raw material commodities) may also be calculated based on ICF recall requirements, for example, according to the types of the raw materials such as text, audio, image, and video, vector expression embedding of each raw material commodity is comprehensively calculated in manners such as NLP, image understanding, and video understanding, and then the raw materials that are most similar to each raw material are clustered by embedding, so that the raw material similarity between each raw material commodity is calculated, and subsequently, other raw material commodities that are similar to the target raw material can be quickly determined according to the raw material similarity. Then, a second recommended candidate set for each user may be determined according to the user similarity and the material similarity. In addition, during data recall, there may be multiple recall modes, including interest-based content recall, collaborative filtering recall, algorithm recall, and the like. The finally determined second recommended candidate set can be determined according to the combined results of the three recalling modes. Specifically, the interest content recall means that raw material commodities with different dimensions can be matched for the user based on the interest images of the user. For example, assuming that the user likes all-cotton, the latest/hottest recall of content can be selected from all-cotton content as the second recommended candidate set for the user, which is very intuitive and interpretable. One-level or multi-level indexes are formed according to different dimensions to form multi-path interest recalls, and because interest points of all dimensions in the user portrait have weights, the number of corresponding recalls can be adjusted in a self-adaptive mode according to the weights. The collaborative filtering recall refers to recommending some raw material commodities for the user based on the user similarity and the raw material similarity. Based on the User similarity, the recommendation can be simply understood as that "the materials which are watched by the users who watch the same raw materials with you" are recommended to you ", namely User collectivity Filtering-UCF, and based on the principle of the raw material similarity, the recommendation can be simply understood as that" the similar raw materials which are watched by the users are recommended to you ", namely Item collectivity Filtering-ICF, so that the basis of the recall mode is the process of finding the similar users or the similar raw materials. The algorithm recall means that in an off-line state, a model is trained through operation records, context of a user, raw material and raw material characteristics related in the operation records and sample data, the trained model is synchronized to a prediction service, and then the user id and the user context can be acquired on line and transmitted to the prediction service to generate user characteristics and context characteristics. And calculating the score of each raw material commodity by combining the raw material characteristics and the model, and after sorting, taking topN as a recall set and returning the recall set to a recommendation engine. The context may include information such as a user operation time series, a commodity creation time series, and the like.
The algorithm needs data, the data needs characteristics, the characteristics are the most important part in model training data, the essence of each algorithm is to fit a probability distribution function which is closest to the real situation according to the existing sample distribution, the parameters needing to be fitted are the weight of each characteristic, and the interest degree (click probability) of a user on the raw material under the characteristics of a specific user or the raw material can be obtained through the function. User characteristics including context characteristics are counted according to a user offline behavior log, a user portrait is similar, the user attribute category, each dimension preference, statistics on each behavior at the moment are assumed, the matching degree and the like related to the factory portrait attribute are processed by using characteristic engineering and stored in a hive table in a structured mode, similarly, the attribute category characteristics of raw materials are obtained in a resource pool, the statistics type information of each dimension of the raw materials is obtained through a client log and stored in the hive table, then characteristic data in the hive is integrated through a timing task, and therefore the user and raw material dimension characteristics which are calculated offline are stored in the hive table and are updated to a redis supply line to be obtained in real time.
Recalls based on these bases can be used as candidate sets. The final recommendation set recommendation of the user can be determined according to the first recommendation candidate set, the second recommendation candidate set and the third recommendation candidate set determined in the above manner, and then the raw material commodities included in the final recommendation set can be determined as recommended raw material commodities for the user. The system can also carry out work such as sorting, reordering and the like backwards on the recommendation set, and realize a plurality of subsequent recommendation functions on the basis of the work. Specifically, in one embodiment, determining the raw material items included in the final recommendation set as recommended raw material items for the user further includes: sorting the raw material commodities included in the final recommendation set according to the sequence of the recommendation degrees from high to low; adjusting the sequence and recommended quantity of raw material commodities belonging to the same type; selecting a preset number of raw material commodities with the serial numbers in front to determine the raw material commodities as recommended raw material commodities of the user. At least one of behavioral data of the user, material characteristics of the respective textile materials, context information may be input to the predictive model to determine a third recommended candidate set for the user. That is, the input data may be analyzed by the predictive model to determine a third recommended candidate set for each user for the raw material commodities that the user may be interested in. A final recommendation set for the user may be determined according to the first recommendation candidate set, the second recommendation candidate set, and the third recommendation candidate set, and the material commodities included in the final recommendation set may be determined as recommended material commodities for the user.
In one embodiment, the method further comprises: acquiring a real-time operation log of a user; determining at least one of recommended exposure rate and exposure click rate of the raw material commodity according to the operated log; and adjusting the recommended raw material commodity aiming at the user according to the recommended exposure rate and the exposure click rate.
The real-time stream is used for acquiring effective information from the log in real time, and for applications with high access frequency of users, which belong to necessities, real-time behaviors of the users and real-time statistical information of raw materials can be fed back and processed by using a stream type calculation tool flink. Generally, offline image data which can be generated from a data warehouse is a day-level or hour-level timing task, and some scenes which need to feed back user behaviors in real time cannot meet requirements, for example, data which are just exposed by a user need to be filtered or the right of the user is reduced, raw material commodities which are just clicked can be pushed to similar articles when the raw material commodities are recommended, some negative feedback information needs to be recorded and takes effect in real time, and the information can be quickly acquired by calculating in real time through user logs which are collected in real time according to different requirement logics and fed back to a system to quickly make changes. Specifically, through collecting the operation logs of the user on the APP in real time, and through a big data real-time calculation engine flink and the like, the recommended exposure rate, the exposure click rate and other indexes of the raw material commodities on the APP are calculated. The recommended exposure rate is the number of raw material commodities that the user can actually see after recommendation/the number of raw material commodities recommended in real time. The exposure click rate is the actual number of material commodities clicked/the actual number of material commodities that the user can actually see after recommendation. If the recommended exposure rate is high, the user is willing to automatically recommend the raw material commodity list to the user by a direct brushing system, so that the user can continuously brush the list and expose different raw material commodities all the time. If the exposure click rate is high, the user is willing to not only refresh the raw material commodity list recommended to the user by the system, but also click in the detailed content of the specific raw material commodity, such as the abstract on the list. And generally outputting the real-time calculation result to different redis keys according to different functions for online use. The calculation targets of the raw material side are similar, for example, the real-time conditions of data such as exposure, click, praise and the like of the raw material are similar, and particularly, after the new raw material is subjected to cold start exposure, the effect trend of the raw material can be reflected quickly, the quality of the raw material is judged, and the subsequent recommendation strategy is influenced. Real-time calculations of the feedstock also exist in redis.
In one embodiment, the method further comprises: performing feature extraction on the behavior data to determine corresponding user features; determining the commodity characteristics of the raw material commodities corresponding to each operation behavior contained in the acquired behavior data; determining the characteristic weight of each user characteristic according to the commodity characteristics and the behavior data; and adjusting the recommended raw material commodity aiming at the user through the characteristic weight.
The three recalls are combined for use, a candidate set is formed by multiple recalls, data such as user exposure, clicking or negative feedback and the like are filtered according to business rules in the recalling process, meanwhile, the recalling aspect also needs to give consideration to user interest and diversity exploration, otherwise, the user possibly gets more and more contents which are detailed and watched, the operation space of subsequent fine-ranking is small, and the contents which are unknown and preferred by the user cannot be pushed out. Therefore, the candidate set categories need to be diversified in the recall stage, the principle of collaborative filtering can generate certain diversity but also make a turn in a similar circle, and recall can be explored through more categories or similar categories. The user experience can be optimized through rearrangement, the control amount or the break-up of the similar content also needs to be adjusted in a strategy mode according to the user behavior, and the measurement of the break-up of the next control amount is determined according to a few continuous points brushed or no continuous points of the user. Meanwhile, the interest exploration needs to be performed by adjusting the content diversity, especially for users with insufficient portraits and behaviors. Meanwhile, the operation business also needs to adjust the ranking, such as setting the top of the related policy content of the textile industry, weighting the hot content, and increasing or decreasing the right of the operation content. And (4) all the reordered raw materials are packaged and sent to be delivered, topN truncation is carried out according to the quantity of the returned results called by the front end, and the cut-off results are displayed for a user. Therefore, the recommended exposure rate and the exposure click rate of the raw material commodity can be determined according to the real-time operation log of the user, and the recommended raw material commodity for the user can be adjusted according to the recommended exposure rate and the exposure click rate.
In the present application, an online prediction function may also be provided. And the prediction inference service can be provided in real time by using a prediction service form on line, and the raw material id, the user id and the request context information of the candidate set to be sorted are transmitted to the prediction service through a recommendation engine interface. The prediction service is also divided into modules of feature extraction, raw material scoring and sorting and the like. And (3) extracting characteristics from the characteristic library on line through the transmitted raw material id and the user id, obtaining characteristic information of all candidate sets by combining the context characteristics, and calculating the score of each raw material through the weight of each characteristic in the model. In the process, the extracted characteristic id is kept consistent with the characteristic id in the trained model, and meanwhile, the system performance is improved in the processes of extracting and scoring the characteristic of each raw material in a parallelization mode. The trained model is synchronized to the online prediction service machine in a timing mode through an offline training process.
According to the method for recommending the textile raw materials, the factory capacity big data of the textile industry is utilized, the requirements of individual users and authenticated enterprise users on more accurate raw material matching recommendation are met according to the matching relations of user images, raw material commodity images and the like, more intelligent raw material commodity recommendation is achieved, and the recommendation accuracy is higher.
Embodiments of the present application provide a storage medium having a program stored thereon, which when executed by a processor, implements the above-described textile material recommendation method.
The embodiment of the application provides a processor, and the processor is used for executing a program, wherein the program executes the method for recommending the textile raw materials during running.
In one embodiment, a textile material recommendation device is provided that includes the processor described above.
The processor comprises a kernel, and the kernel calls the corresponding program unit from the memory. One or more than one inner core can be set, and the method for recommending the textile raw materials is realized by adjusting the parameters of the inner core.
The memory may include volatile memory in a computer readable medium, Random Access Memory (RAM) and/or nonvolatile memory such as Read Only Memory (ROM) or flash memory (flash RAM), and the memory includes at least one memory chip.
In one embodiment, a computer device is provided, which may be a server, the internal structure of which may be as shown in fig. 3. The computer device includes a processor a01, a network interface a02, a memory (not shown), and a database (not shown) connected by a system bus. Wherein processor a01 of the computer device is used to provide computing and control capabilities. The memory of the computer device comprises an internal memory a03 and a non-volatile storage medium a 04. The non-volatile storage medium a04 stores an operating system B01, a computer program B02, and a database (not shown in the figure). The internal memory a03 provides an environment for the operation of the operating system B01 and the computer program B02 in the nonvolatile storage medium a 04. The network interface a02 of the computer device is used for communication with an external terminal through a network connection. The computer program B02 is executed by the processor a01 to implement a textile material recommendation method.
Those skilled in the art will appreciate that the architecture shown in fig. 3 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
The embodiment of the application provides equipment, the equipment comprises a processor, a memory and a program which is stored on the memory and can run on the processor, and the following steps are realized when the processor executes the program: acquiring behavior data of a user; determining the interest degree of the users in each dimension aiming at the textile raw material commodities and the user similarity among the users according to the behavior data; determining a first recommendation candidate set for the user according to the interest degree; determining the raw material similarity between raw material commodities; determining a second recommended candidate set of the user according to the user similarity and the raw material similarity; inputting at least one of the behavior data of the user, the raw material characteristics of each textile raw material and the context information into a prediction model to determine a third recommended candidate set of the user; determining a final recommendation set of the user according to the first recommendation candidate set, the second recommendation candidate set and the third recommendation candidate set; and determining the raw material commodities included in the final recommendation set as recommended raw material commodities for the user.
In one embodiment, the behavioral data includes user operational behavior for the raw commodity over a plurality of time periods of different durations; determining the interest degree of the users in the textile raw material commodities and the user similarity among the users according to the behavior data comprises the following steps: acquiring commodity attributes of raw material commodities corresponding to each operation behavior of a user; and determining the interest degree of the user for each raw material commodity according to the operation behaviors and the commodity attributes.
In one embodiment, the item attributes include item keywords; determining the interest degree of the user for each raw material commodity according to the operation behaviors and the commodity attributes comprises the following steps: determining interest scores of the users for the commodity keywords of the raw material commodities aiming at each raw material commodity; ranking the interest scores to determine a preference sequence of the user for the commodity keywords; and determining the interest degree of the user in the keyword dimension for each raw material commodity according to the preference sequence.
In one embodiment, determining the interest score of the user for the item keyword of the raw item comprises: the interest score is determined according to the following formula:
wherein, scorekThe interest score of the user on the keyword k is indicated, i refers to the raw material commodity operated by the user i in the behavior data, a iMeans the action weight, w, corresponding to the raw material commodity of the ith operationiIs the weight of the keyword k to the material commodity of the ith operation, f (t)i) The time decay function shows the interest decay degree of the user on the raw material commodity of the ith operation at the time t.
In one embodiment, determining raw material similarity between raw material commodities comprises: acquiring raw material data of each raw material commodity; determining a characteristic vector of each raw material commodity according to the raw material data; and clustering the feature vectors to determine the raw material similarity between the raw material commodities.
In one embodiment, determining the raw material items included in the final recommendation set as recommended raw material items for the user further comprises: sorting the raw material commodities included in the final recommendation set according to the sequence of the recommendation degrees from high to low; adjusting the sequence and recommended quantity of raw material commodities belonging to the same type; selecting a preset number of raw material commodities with the serial numbers in front to determine the raw material commodities as recommended raw material commodities of the user.
In one embodiment, the method further comprises: performing feature extraction on the behavior data to determine corresponding user features; determining the commodity characteristics of the raw material commodities corresponding to each operation behavior contained in the acquired behavior data; determining the characteristic weight of each user characteristic according to the commodity characteristics and the behavior data; and adjusting the recommended raw material commodity aiming at the user through the characteristic weight.
In one embodiment, the method further comprises: acquiring a real-time operation log of a user; determining at least one of recommended exposure rate and exposure click rate of the raw material commodity according to the operation log; and adjusting the recommended raw material commodity aiming at the user according to the recommended exposure rate and the exposure click rate.
The present application further provides a computer program product adapted to perform a program for initializing the following method steps when executed on a data processing device: acquiring behavior data of a user; determining the interest degree of the users in each dimension aiming at the textile raw material commodities and the user similarity among the users according to the behavior data; determining a first recommendation candidate set for the user according to the interest degree; determining the raw material similarity between raw material commodities; determining a second recommended candidate set of the user according to the user similarity and the raw material similarity; inputting at least one of the behavior data of the user, the raw material characteristics of each textile raw material and the context information into a prediction model to determine a third recommended candidate set of the user; determining a final recommendation set of the user according to the first recommendation candidate set, the second recommendation candidate set and the third recommendation candidate set; and determining the raw material commodities included in the final recommendation set as recommended raw material commodities for the user.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). The memory is an example of a computer-readable medium.
Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in the process, method, article, or apparatus that comprises the element.
The above are merely examples of the present application and are not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.
Claims (10)
1. A method of textile material recommendation, the method comprising:
acquiring behavior data of a user;
determining the interest degree of the user in each dimension for the textile raw material commodities and the user similarity among the users according to the behavior data;
determining a first recommended candidate set for the user according to the interest degree;
determining raw material similarity between the raw material commodities;
determining a second recommended candidate set of the user according to the user similarity and the raw material similarity;
inputting at least one of behavioral data of the user, material characteristics of each textile material, and context information into a predictive model to determine a third recommended candidate set of the user;
determining a final recommendation set for the user according to the first, second, and third recommendation candidate sets;
Determining the raw material commodities included in the final recommendation set as recommended raw material commodities for the user.
2. The method of claim 1, wherein the behavioral data comprises operational behavior of the user with respect to raw material commodities over a plurality of time periods of different durations; the determining the interest degree of the users in the textile raw material commodities and the user similarity among the users according to the behavior data comprises the following steps:
acquiring commodity attributes of the raw material commodities corresponding to each operation behavior of the user;
and determining the interest degree of the user for each raw material commodity according to the operation behaviors and the commodity attributes.
3. The method of claim 2, wherein the merchandise attributes include merchandise keywords; the determining the interest degree of the user for each raw material commodity according to the operation behaviors and the commodity attributes comprises:
for each raw material commodity, determining an interest score of the user for a commodity keyword of the raw material commodity;
ranking the interest scores to determine a preference sequence of the user for commodity keywords;
and determining the interest degree of the user for each raw material commodity in the keyword dimension according to the preference sequence.
4. The method of claim 3, wherein determining the interest score of the user for the item keyword of the raw item comprises: determining the interest score according to the following formula:
wherein, scorekThe interest score of the user on the keyword k is indicated, i refers to the raw material commodity operated by the user i in the behavior data, aiMeans that the raw material commodity of the ith operation corresponds toA behavioral weight of, wiIs the weight of the keyword k to the material commodity of the ith operation, f (t)i) The time decay function shows the interest decay degree of the user on the raw material commodity of the ith operation at the time t.
5. The method of claim 1, wherein the determining feedstock similarities between the feedstock commodities comprises:
acquiring raw material data of each raw material commodity;
determining a characteristic vector of each raw material commodity according to the raw material data;
and clustering the characteristic vectors to determine the raw material similarity among the raw material commodities.
6. The method of claim 1, wherein the determining the raw materials items included in the final recommendation set as recommended raw materials items for the user further comprises:
Sorting the raw material commodities included in the final recommendation set according to the sequence of the recommendation degrees from high to low;
adjusting the sequence and recommended quantity of raw material commodities belonging to the same type;
and selecting a preset number of raw material commodities with the serial numbers in front to determine the raw material commodities as recommended raw material commodities of the user.
7. The method of claim 1, further comprising:
performing feature extraction on the behavior data to determine corresponding user features;
determining and obtaining commodity characteristics of the raw material commodity corresponding to each operation behavior contained in the behavior data;
determining a feature weight of each user feature according to the commodity features and the behavior data;
and adjusting the recommended raw material commodity for the user through the characteristic weight.
8. The method of claim 1, further comprising:
acquiring a real-time operation log of the user;
determining at least one of recommended exposure rate and exposure click rate of the raw material commodity according to the operation log;
and adjusting the recommended raw material commodity for the user according to the recommended exposure rate and the exposure click rate.
9. A processor characterized by being configured to perform the textile material recommendation method according to any one of claims 1 to 8.
10. A textile material recommendation device, characterized in that said device comprises a processor according to claim 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110850576.7A CN113592589B (en) | 2021-07-27 | 2021-07-27 | Textile raw material recommendation method, device and processor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110850576.7A CN113592589B (en) | 2021-07-27 | 2021-07-27 | Textile raw material recommendation method, device and processor |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113592589A true CN113592589A (en) | 2021-11-02 |
| CN113592589B CN113592589B (en) | 2024-03-29 |
Family
ID=78250427
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110850576.7A Active CN113592589B (en) | 2021-07-27 | 2021-07-27 | Textile raw material recommendation method, device and processor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113592589B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114723492A (en) * | 2022-04-15 | 2022-07-08 | 山东浪潮爱购云链信息科技有限公司 | Enterprise portrait generation method and equipment |
| CN115587261A (en) * | 2022-12-09 | 2023-01-10 | 思创数码科技股份有限公司 | Government affair resource catalog recommendation method and system |
Citations (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2011141666A (en) * | 2010-01-06 | 2011-07-21 | Nippon Telegr & Teleph Corp <Ntt> | Device, method and program for recommending interest information |
| JP2013196653A (en) * | 2012-03-22 | 2013-09-30 | Nippon Telegr & Teleph Corp <Ntt> | Interest analysis method, device and program |
| US8549013B1 (en) * | 2010-09-17 | 2013-10-01 | Amazon Technologies, Inc. | Systems and methods for determining interest in an item or category of items |
| US20130317910A1 (en) * | 2012-05-23 | 2013-11-28 | Vufind, Inc. | Systems and Methods for Contextual Recommendations and Predicting User Intent |
| JP2013257668A (en) * | 2012-06-11 | 2013-12-26 | Nippon Telegr & Teleph Corp <Ntt> | Interest analysis method, interest analyzer and program of the same |
| CN104615775A (en) * | 2015-02-26 | 2015-05-13 | 北京奇艺世纪科技有限公司 | User recommendation method and device |
| CN107609960A (en) * | 2017-10-18 | 2018-01-19 | 口碑(上海)信息技术有限公司 | Rationale for the recommendation generation method and device |
| CN107730289A (en) * | 2016-08-11 | 2018-02-23 | 株式会社理光 | A kind of user behavior analysis method and user behavior analysis device |
| CN108550068A (en) * | 2018-04-16 | 2018-09-18 | 南京大学 | A kind of individual commodity recommendation method and system based on user behavior analysis |
| CN109977309A (en) * | 2019-03-21 | 2019-07-05 | 杭州电子科技大学 | Combination point of interest querying method based on multiple key and user preference |
| WO2019141072A1 (en) * | 2018-01-22 | 2019-07-25 | 阿里巴巴集团控股有限公司 | Method, device, and client for recommending store information |
| CN110852852A (en) * | 2019-11-15 | 2020-02-28 | 成都航天科工大数据研究院有限公司 | Industrial Internet product recommendation system and method |
| WO2020175732A1 (en) * | 2019-02-26 | 2020-09-03 | 주식회사 틸투원 | Commercial product recommendation method and device therefor |
| CN111931056A (en) * | 2020-08-17 | 2020-11-13 | 北京小川科技有限公司 | Push content recommendation method and device |
| CN112149003A (en) * | 2020-10-28 | 2020-12-29 | 浙江集享电子商务有限公司 | Commodity community recommendation method and device and computer equipment |
| CN112328881A (en) * | 2020-11-05 | 2021-02-05 | 中国平安人寿保险股份有限公司 | Article recommendation method and device, terminal device and storage medium |
| CN112632359A (en) * | 2020-12-30 | 2021-04-09 | 北京锐安科技有限公司 | Information recommendation method and device, electronic equipment and storage medium |
| CN112785397A (en) * | 2021-03-09 | 2021-05-11 | 中国工商银行股份有限公司 | Product recommendation method, device and storage medium |
-
2021
- 2021-07-27 CN CN202110850576.7A patent/CN113592589B/en active Active
Patent Citations (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2011141666A (en) * | 2010-01-06 | 2011-07-21 | Nippon Telegr & Teleph Corp <Ntt> | Device, method and program for recommending interest information |
| US8549013B1 (en) * | 2010-09-17 | 2013-10-01 | Amazon Technologies, Inc. | Systems and methods for determining interest in an item or category of items |
| JP2013196653A (en) * | 2012-03-22 | 2013-09-30 | Nippon Telegr & Teleph Corp <Ntt> | Interest analysis method, device and program |
| US20130317910A1 (en) * | 2012-05-23 | 2013-11-28 | Vufind, Inc. | Systems and Methods for Contextual Recommendations and Predicting User Intent |
| JP2013257668A (en) * | 2012-06-11 | 2013-12-26 | Nippon Telegr & Teleph Corp <Ntt> | Interest analysis method, interest analyzer and program of the same |
| CN104615775A (en) * | 2015-02-26 | 2015-05-13 | 北京奇艺世纪科技有限公司 | User recommendation method and device |
| CN107730289A (en) * | 2016-08-11 | 2018-02-23 | 株式会社理光 | A kind of user behavior analysis method and user behavior analysis device |
| CN107609960A (en) * | 2017-10-18 | 2018-01-19 | 口碑(上海)信息技术有限公司 | Rationale for the recommendation generation method and device |
| WO2019141072A1 (en) * | 2018-01-22 | 2019-07-25 | 阿里巴巴集团控股有限公司 | Method, device, and client for recommending store information |
| CN108550068A (en) * | 2018-04-16 | 2018-09-18 | 南京大学 | A kind of individual commodity recommendation method and system based on user behavior analysis |
| WO2020175732A1 (en) * | 2019-02-26 | 2020-09-03 | 주식회사 틸투원 | Commercial product recommendation method and device therefor |
| CN109977309A (en) * | 2019-03-21 | 2019-07-05 | 杭州电子科技大学 | Combination point of interest querying method based on multiple key and user preference |
| CN110852852A (en) * | 2019-11-15 | 2020-02-28 | 成都航天科工大数据研究院有限公司 | Industrial Internet product recommendation system and method |
| CN111931056A (en) * | 2020-08-17 | 2020-11-13 | 北京小川科技有限公司 | Push content recommendation method and device |
| CN112149003A (en) * | 2020-10-28 | 2020-12-29 | 浙江集享电子商务有限公司 | Commodity community recommendation method and device and computer equipment |
| CN112328881A (en) * | 2020-11-05 | 2021-02-05 | 中国平安人寿保险股份有限公司 | Article recommendation method and device, terminal device and storage medium |
| CN112632359A (en) * | 2020-12-30 | 2021-04-09 | 北京锐安科技有限公司 | Information recommendation method and device, electronic equipment and storage medium |
| CN112785397A (en) * | 2021-03-09 | 2021-05-11 | 中国工商银行股份有限公司 | Product recommendation method, device and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 王林;刘继源;马安进;: "基于兴趣衰减的个性化排序算法", 计算机工程, no. 09 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114723492A (en) * | 2022-04-15 | 2022-07-08 | 山东浪潮爱购云链信息科技有限公司 | Enterprise portrait generation method and equipment |
| CN115587261A (en) * | 2022-12-09 | 2023-01-10 | 思创数码科技股份有限公司 | Government affair resource catalog recommendation method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113592589B (en) | 2024-03-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110503531B (en) | Dynamic social scene recommendation method based on time sequence perception | |
| Liu et al. | Online evolutionary collaborative filtering | |
| Li et al. | A contextual-bandit approach to personalized news article recommendation | |
| Karatzoglou et al. | Multiverse recommendation: n-dimensional tensor factorization for context-aware collaborative filtering | |
| US20230066853A1 (en) | Method and apparatus for training information prediction models, method and apparatus for predicting information, and storage medium and device thereof | |
| WO2014148948A1 (en) | Method for selecting valid variants in search and recommendation systems (variants) | |
| CN110781409A (en) | Article recommendation method based on collaborative filtering | |
| CN113592589B (en) | Textile raw material recommendation method, device and processor | |
| CN110175895A (en) | A kind of item recommendation method and device | |
| CN110688565B (en) | Next item recommendation method based on multidimensional Hox process and attention mechanism | |
| CN111858969B (en) | Multimedia data recommendation method, device, computer equipment and storage medium | |
| WO2012034606A2 (en) | Multiverse recommendation method for context-aware collaborative filtering | |
| Kannikaklang et al. | A hybrid recommender system for improving rating prediction of movie recommendation | |
| Guan et al. | Enhanced SVD for collaborative filtering | |
| CN114579843A (en) | Personalized search method and device, computer equipment and storage medium | |
| CN116501979A (en) | Information recommendation method, information recommendation device, computer equipment and computer readable storage medium | |
| KR20200145346A (en) | A program for the product or content recommendation service | |
| EP2715570A1 (en) | Determining matching degrees between information categories and displayed information | |
| KR20230061222A (en) | Method for Content Recommendation in multi-type User Environment Based on Machine Learning | |
| CN113536085B (en) | Method and system for scheduling subject term search crawlers based on combined prediction method | |
| Trinks | A classification of real time analytics methods. an outlook for the use within the smart factory | |
| CN115222177A (en) | Service data processing method and device, computer equipment and storage medium | |
| CN108197335B (en) | Personalized query recommendation method and device based on user behaviors | |
| CN113627967A (en) | Behavior data processing method, device, equipment and storage medium | |
| KR20200145292A (en) | A computer-readable recording medium on which a program for executing a user preference prediction method based on collaborative filtering is recorded |
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 |