CN117874347A - Content recommendation technology based on business characteristics - Google Patents
Content recommendation technology based on business characteristics Download PDFInfo
- Publication number
- CN117874347A CN117874347A CN202410056798.5A CN202410056798A CN117874347A CN 117874347 A CN117874347 A CN 117874347A CN 202410056798 A CN202410056798 A CN 202410056798A CN 117874347 A CN117874347 A CN 117874347A
- Authority
- CN
- China
- Prior art keywords
- recommendation
- user
- similarity
- algorithm
- users
- 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
- 238000005516 engineering process Methods 0.000 title claims abstract description 14
- 238000004364 calculation method Methods 0.000 claims abstract description 15
- 238000011156 evaluation Methods 0.000 claims abstract description 13
- 238000007781 pre-processing Methods 0.000 claims abstract description 12
- 238000013480 data collection Methods 0.000 claims abstract description 8
- 238000000605 extraction Methods 0.000 claims abstract description 7
- 238000005457 optimization Methods 0.000 claims abstract description 7
- 238000001914 filtration Methods 0.000 claims description 32
- 239000013598 vector Substances 0.000 claims description 32
- 230000006399 behavior Effects 0.000 claims description 27
- 238000000034 method Methods 0.000 claims description 26
- 230000008569 process Effects 0.000 claims description 10
- 230000003044 adaptive effect Effects 0.000 claims description 3
- 238000000354 decomposition reaction Methods 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 3
- 230000000694 effects Effects 0.000 description 3
- 238000007418 data mining Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000008713 feedback mechanism Effects 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 230000003997 social interaction Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9536—Search customisation based on social or collaborative filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/217—Validation; Performance evaluation; Active pattern learning techniques
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention relates to the field of content recommendation, in particular to a content recommendation technology based on service characteristics, which comprises the steps of data collection and preprocessing, characteristic extraction, similarity calculation, algorithm recommendation, evaluation and optimization, wherein the data collection and preprocessing collect user behavior data and article information data and preprocess the user behavior data and the article information data.
Description
Technical Field
The invention relates to the field of content recommendation, in particular to a content recommendation technology based on service characteristics.
Background
The core technologies of the intelligent recommendation system comprise machine learning, data mining, natural language processing and recommendation algorithms, and the development and application of the technologies enable the intelligent recommendation system to better understand interests and demands of users, so that personalized and accurate recommendation results are provided.
The traditional collaborative filtering-based recommendation system has poor effect under the condition of processing new users or lacking user behavior data, because of lacking enough user data, the system is difficult to accurately predict interests and demands of the new users, the collaborative filtering-based recommendation system mainly depends on similarities among users, and features and preferences of individual users are ignored, so that recommendation results lack of individuation, specific demands of users cannot be met, and when sparse data and large-scale data are processed by a collaborative filtering algorithm, calculation complexity is high, and the real-time performance and expandability of the recommendation system are insufficient, so that the demands of high-concurrency and large-scale users are difficult to deal with.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides a content recommendation technology based on service characteristics.
The technical scheme adopted for solving the technical problems is as follows: a content recommendation technology based on business features comprises the steps of data collection and preprocessing, feature extraction, similarity calculation, algorithm recommendation, evaluation and optimization, and the method comprises the following steps:
s1: data collection and preprocessing, namely collecting user behavior data and article information data, preprocessing the user behavior data, de-duplicating, filtering and standardizing the user behavior data, and extracting and encoding the characteristics of the article information data;
s2: extracting characteristics, namely extracting useful characteristics from user behavior data and article information data, wherein the characteristics are used for describing the attributes of users and articles, the characteristics comprise the historical behaviors of the users, the content characteristics of the articles and the attributes of the users and the articles, and characteristic engineering adopts TF-IDF, word2Vec and One-Hot coding;
s3: calculating the similarity between the user and the object according to the characteristics of the user and the object, wherein various methods such as cosine similarity, euclidean distance and Pearson correlation coefficient are used for calculating the similarity between the users by using the cosine similarity or the similarity between the objects by using the content similarity;
s4: algorithm recommendation, based on the result of similarity calculation, generating personalized recommendation for the user by using a recommendation algorithm, wherein the recommendation algorithm comprises collaborative filtering, content filtering and mixed recommendation, and the following formula of an example recommendation algorithm is as follows:
wherein NN represents a set of users similar to the target user v, the similarity (u, v) represents the similarity between the user uu and the user v, the user score (u) represents the score of the user uu on the item, and the recommendation algorithm based on matrix decomposition:
recommendation algorithm based on Contents filtration:
s5: and evaluating and optimizing the generated recommendation result, evaluating the performance of the recommendation system by using some evaluation indexes such as accuracy, recall rate and coverage rate, and optimizing and adjusting a recommendation algorithm according to the evaluation result so as to improve the recommendation accuracy and user satisfaction.
Specifically, the collaborative filtering algorithm comprises a collaborative filtering algorithm based on a user and a collaborative filtering algorithm based on an article;
collaborative filtering algorithm based on user: the algorithm recommends according to the similarity between users, the similarity between users is measured by calculating Euclidean distance and cosine similarity indexes between users, and in the recommendation process, favorite articles are recommended to target users according to the behaviors and preferences of other users similar to the target users;
collaborative filtering algorithm based on articles: the algorithm recommends according to the similarity between the articles, the similarity between the articles is measured by calculating the correlation between the articles and the degree index which is commonly liked by users, and other articles similar to the articles are found to be recommended according to the articles liked by target users in the recommendation process.
Specifically, the TF-IDF algorithm: the algorithm is used for calculating the importance of each word in the text, and measuring the importance of the word by calculating word frequency and inverse document frequency, so that feature vectors are generated for the text, and in the recommendation process, the content with similar feature vectors is recommended according to the interests and the favorites of a target user;
and obtaining importance scores of each word for each document by calculating TF-IDF values of all words in the document set, thereby being used for text retrieval and keyword extraction tasks.
Specifically, the similarity calculation uses a cosine similarity algorithm to calculate the similarity between the interest vector of the user and the feature vector of the content, so as to recommend the interest vector;
the degree of similarity between two vectors is evaluated by calculating their cosine similarity.
In particular, the algorithm recommends that to better understand the needs and context of the user, the system incorporates context-aware and adaptive recommendations, including taking into account the user's current environment, time, geographic location factors, and the user's real-time behavior and feedback, and the system uses sensor data, location information, timestamps to incorporate the context information into the recommendation algorithm, providing more accurate and practical recommendation results.
Specifically, the evaluation and optimization includes an accuracy rate and a recall rate, the accuracy rate represents the proportion of the truly related items in the recommendation list to the recommendation list, and the recall rate represents the proportion of the truly related items in the recommendation list to all related items.
The invention has the beneficial effects that:
(1) According to the content recommendation technology based on the service characteristics, through an advanced machine learning algorithm and a data mining technology and by combining multidimensional user data and content information, interests and requirements of users can be more accurately understood, personalized and accurate recommendation results are provided, and the system can better adapt to requirements of different users by overcoming the problems of new users and cold start.
(2) According to the content recommendation technology based on the business characteristics, the personal information, the historical behavior and the multi-dimensional data of the social network data of the user are comprehensively considered, a more targeted recommendation result is provided for the user, and the system can better meet specific requirements of the user and provide personalized recommendation experience by deeply mining interests and preferences of the user.
(3) According to the content recommendation technology based on the service characteristics, the personalized recommendation result of the user can be updated timely by introducing the real-time recommendation and feedback mechanism, dynamic adjustment is performed according to the feedback of the user, and the calculation efficiency and the response speed of a recommendation system are improved by optimizing an algorithm and a system architecture, so that the requirements of high-concurrency and large-scale users are met.
Drawings
The invention will be further described with reference to the drawings and examples.
Fig. 1 is a flowchart of a content recommendation technique based on service features provided in the present invention.
Detailed Description
The invention is further described in connection with the following detailed description in order to make the technical means, the creation characteristics, the achievement of the purpose and the effect of the invention easy to understand.
In embodiment 1, as shown in fig. 1, the content recommendation technology based on service features of the present invention includes the steps of data collection and preprocessing, feature extraction, similarity calculation, algorithm recommendation, evaluation and optimization, and the method includes the following steps:
s1: data collection and preprocessing, namely collecting user behavior data and article information data, preprocessing the user behavior data, de-duplicating, filtering and standardizing the user behavior data, and extracting and encoding the characteristics of the article information data;
collecting behavior data of a user, clicking records, purchase histories and scoring;
collecting social network data, friend relation and social interaction of a user;
preprocessing data, removing noise and processing missing values;
s2: extracting characteristics, namely extracting useful characteristics from user behavior data and article information data, wherein the characteristics are used for describing the attributes of users and articles, the characteristics comprise the historical behaviors of the users, the content characteristics of the articles and the attributes of the users and the articles, and characteristic engineering adopts TF-IDF, word2Vec and One-Hot coding;
extracting characteristics from behavior data of a user, and selecting interest preferences, purchasing preferences and liveness of the user;
extracting features from social network data of a user, and social influence and social circle of the user;
s3: calculating the similarity between the user and the object according to the characteristics of the user and the object, wherein various methods such as cosine similarity, euclidean distance and Pearson correlation coefficient are used for calculating the similarity between the users by using the cosine similarity or the similarity between the objects by using the content similarity;
s4: algorithm recommendation, based on the result of similarity calculation, generating personalized recommendation for the user by using a recommendation algorithm, wherein the recommendation algorithm comprises collaborative filtering, content filtering and mixed recommendation, and the following formula of an example recommendation algorithm is as follows:
wherein NN represents a set of users similar to the target user v, the similarity (u, v) represents the similarity between the user uu and the user v, the user score (u) represents the score of the user uu on the item, and the recommendation algorithm based on matrix decomposition:
recommendation algorithm based on Contents filtration:
s5: evaluating and optimizing the generated recommendation result, evaluating the performance of the recommendation system by using some evaluation indexes such as accuracy, recall rate and coverage rate, and optimizing and adjusting a recommendation algorithm according to the evaluation result so as to improve the recommendation accuracy and user satisfaction;
generating a recommendation list based on the user by using a collaborative filtering algorithm based on the similarity and behavior data of the user;
generating a recommendation list based on social relations by using a social recommendation algorithm based on social network data of the user;
weighting and fusing the recommendation list based on the user and the recommendation list based on the social relationship to generate a final mixed recommendation list;
filtering out items that the user has purchased or is not interested in;
and sequencing the recommendation results according to the characteristics of the recommended articles and the interest preference of the user so as to provide personalized recommendation sequences.
Specifically, the collaborative filtering algorithm comprises a collaborative filtering algorithm based on a user and a collaborative filtering algorithm based on an article;
collaborative filtering algorithm based on user: the algorithm recommends according to the similarity between users, the similarity between users is measured by calculating Euclidean distance and cosine similarity indexes between users, and in the recommendation process, favorite articles are recommended to target users according to the behaviors and preferences of other users similar to the target users;
the formula: similarity (u, v) =cosine_similarity (Ru, rv) =dot_product (Ru, rv)/(norm (Ru) ×norm (Rv))
predicted_rating(u，i)＝mean_rating(u)+sum(similarity(u，v)*(rating(v，i)-mean_rating(v)))/sum(similarity(u，v))；
Collaborative filtering algorithm based on articles: the algorithm recommends according to the similarity between the articles, the similarity between the articles is measured by calculating the correlation between the articles and the degree index which is commonly liked by users, and other articles similar to the articles are found to be recommended according to the articles liked by target users in the recommendation process;
the formula: similarity (i, j) =cosine_similarity (Ri, rj) =dot_product (Ri, rj)/(norm (Ri) ×norm (Rj))
predicted_rating(u，i)＝sum(similarity(i，j)*rating(u，j))/sum(similarity(i，j))。
Specifically, the TF-IDF algorithm: the algorithm is used for calculating the importance of each word in the text, and the importance of the words is measured by calculating word frequency and inverse document frequency, so that feature vectors are generated for the text, in the recommendation process, the content with similar feature vectors is recommended according to the interests and the favorites of a target user,
the formula: TF (t) = (number of times word t occurs in document)/(total number of words in document)
IDF (t) =log ((total number of documents in document set)/(number of documents +1) containing word t), TF-IDF (t, d) =tf (t, d) ×idf (t), where t represents a word, d represents a document,
the larger the TF-IDF value, the higher the importance of the word in the document, and when the word frequency in the current document is high (TF high) and the prevalence in the whole document set is low (IDF low), the TF-IDF value will be higher,
and obtaining importance scores of each word for each document by calculating TF-IDF values of all words in the document set, thereby being used for text retrieval and keyword extraction tasks.
Specifically, the similarity calculation uses a cosine similarity algorithm to calculate the similarity between the interest vector of the user and the feature vector of the content, so as to recommend the interest vector;
the formula: where A and B are two vectors, dot_product (A, B) represents the dot product (inner product) of vector A and vector B, norm (A) represents the norm (length) of vector A, norm (B) represents the norm of vector B;
cosine_similarity(A，B)＝dot_product(A，B)/(norm(A)*norm(B))
calculating a dot product (inner product) of the vector A and the vector B, wherein the dot product is a result obtained by multiplying elements at corresponding positions of the two vectors and adding the products;
dot_product(A，B)＝A[0]*B[0]+A[1]*B[1]+...+A[n]*B[n]
calculating norms (lengths) of the vector A and the vector B, wherein the norms are obtained by adding squares of each element of the vector and taking square roots;
norm(A)＝sqrt(A[0]^2+A[1]^2+...+A[n]^2)
norm(B)＝sqrt(B[0]^2+B[1]^2+...+B[n]^2)
dividing the dot product by the product of the norms to obtain a cosine similarity value;
cosine_similarity(A，B)＝dot_product(A，B)/(norm(A)*norm(B))
the cosine similarity has a value ranging between [ -1,1], the closer the value is to 1 the more similar the two vectors are, the closer the value is to-1 the more dissimilar the two vectors are, the value is 0 the two vectors are orthogonal (no similarity),
the degree of similarity between two vectors is evaluated by calculating their cosine similarity.
In particular, the algorithm recommends that to better understand the needs and context of the user, the system incorporates context-aware and adaptive recommendations, including taking into account the user's current environment, time, geographic location factors, and the user's real-time behavior and feedback, and the system uses sensor data, location information, timestamps to incorporate the context information into the recommendation algorithm, providing more accurate and practical recommendation results.
Specifically, the evaluation and optimization includes an accuracy rate and a recall rate, the accuracy rate represents the proportion of the truly related items in the recommendation list to the recommendation list, the recall rate represents the proportion of the truly related items in the recommendation list to all related items, and the calculation formula is as follows:
accuracy = recommended number of related items/recommended total number of items;
recall rate, recall rate represent recommendation list really relevant article account for all relevant article proportion, the computational formula is: recall = recommended number of related items/number of all related items;
the F1 value, the F1 value comprehensively considers the accuracy and the recall rate, is a harmonic average value of the accuracy and the recall rate, and has the calculation formula:
f1 value = 2 (accuracy rate recall)/(accuracy rate + recall).
In embodiment 2, there are some alternatives to the real-time recommendation algorithm in embodiment 1, in which the method uses offline computing to process the historical data of the user, generates recommendation results in advance and stores the recommendation results in the database, and then updates the recommendation results in the database in real time according to the real-time behavior of the user, and returns to the interface of the user, so that the method can reduce the computation complexity of real-time recommendation, but the recommendation effect is relatively low, and meanwhile, there may be problems such as data delay.
Example 3 there are some alternatives to the real-time recommendation algorithm in example 1, where there is a more common online learning model: under the method, the system can continuously learn and adjust the model from the real-time feedback of the user, and the method for generating the real-time recommendation result has the advantages of realizing more accurate real-time recommendation, having no problems of data delay and the like, but needing to update the online model regularly, and having longer response time of each request due to larger calculation amount.
The foregoing has shown and described the basic principles, principal features and advantages of the invention. It will be understood by those skilled in the art that the present invention is not limited to the foregoing examples, and that the foregoing description and description are merely illustrative of the principles of this invention, and various changes and modifications may be made without departing from the spirit and scope of the invention, which is defined in the appended claims. The scope of the invention is defined by the appended claims and their equivalents.
Claims (6)
1. The content recommendation technology based on the service features is characterized by comprising the steps of data collection and preprocessing, feature extraction, similarity calculation, algorithm recommendation, evaluation and optimization, and the method comprises the following steps:
s1: data collection and preprocessing, namely collecting user behavior data and article information data, preprocessing the user behavior data, de-duplicating, filtering and standardizing the user behavior data, and extracting and encoding the characteristics of the article information data;
s2: extracting characteristics, namely extracting useful characteristics from user behavior data and article information data, wherein the characteristics are used for describing the attributes of users and articles, the characteristics comprise the historical behaviors of the users, the content characteristics of the articles and the attributes of the users and the articles, and the characteristic engineering adopts TF-IDF, word2Vec and One-Hot coding;
s3: calculating similarity, namely calculating the similarity between the user and the object according to the characteristics of the user and the object, wherein various methods are used for calculating the similarity, and cosine similarity, euclidean distance and Pearson correlation coefficient are used for calculating the similarity between the users or content similarity is used for calculating the similarity between the objects;
s4: algorithm recommendation, based on the result of similarity calculation, generating personalized recommendation for the user by using a recommendation algorithm, wherein the recommendation algorithm comprises collaborative filtering, content filtering and mixed recommendation, and the following formula of an example recommendation algorithm is as follows:
wherein NN represents a set of users similar to the target user v, the similarity (u, v) represents the similarity between the user uu and the user v, the user score (u) represents the score of the user uu on the item, and the recommendation algorithm based on matrix decomposition:
recommendation algorithm based on Contents filtration:
s5: and evaluating and optimizing the generated recommendation result, evaluating the performance of the recommendation system by using a plurality of evaluation indexes including accuracy, recall rate and coverage rate, and optimizing and adjusting a recommendation algorithm according to the evaluation result so as to improve the recommendation accuracy and user satisfaction.
2. The business feature based content recommendation technique of claim 1, wherein: the collaborative filtering algorithm comprises a collaborative filtering algorithm based on a user and a collaborative filtering algorithm based on an article;
collaborative filtering algorithm based on user: the algorithm recommends according to the similarity between users, the similarity between users is measured by calculating Euclidean distance and cosine similarity indexes between users, and in the recommendation process, favorite articles are recommended to target users according to the behaviors and preferences of other users similar to the target users;
collaborative filtering algorithm based on articles: the algorithm recommends according to the similarity between the articles, the similarity between the articles is measured by calculating the correlation between the articles and the degree index which is commonly liked by users, and other articles similar to the articles are found to be recommended according to the articles liked by target users in the recommendation process.
3. The content recommendation technique for a basic service feature according to claim 1, wherein: the TF-IDF algorithm: the algorithm is used for calculating the importance of each word in the text, and measuring the importance of the word by calculating word frequency and inverse document frequency, so that feature vectors are generated for the text, and in the recommendation process, the content with similar feature vectors is recommended according to the interests and the favorites of a target user;
and obtaining importance scores of each word for each document by calculating TF-IDF values of all words in the document set, thereby being used for text retrieval and keyword extraction tasks.
4. The business feature based content recommendation technique of claim 1, wherein: the similarity calculation uses a cosine similarity algorithm to calculate the similarity between the interest vector of the user and the characteristic vector of the content, so as to recommend the interest vector;
the degree of similarity between two vectors is evaluated by calculating their cosine similarity.
5. The business feature based content recommendation technique of claim 1, wherein: the algorithm recommends to better understand the needs and context of the user, the system introduces context-aware and adaptive recommendations, including consideration of the user's current environment, time, geographic location factors, and real-time behavior and feedback of the user, and the system uses sensor data, location information, timestamps to incorporate context information into the recommendation algorithm, providing more accurate and practical recommendation results.
6. The business feature based content recommendation technique of claim 1, wherein: the evaluation and optimization comprises accuracy and recall accuracy, wherein the accuracy represents the proportion of the truly related items in the recommendation list to the recommendation list, and the recall represents the proportion of the truly related items in the recommendation list to all related items.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410056798.5A CN117874347A (en) | 2024-01-15 | 2024-01-15 | Content recommendation technology based on business characteristics |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410056798.5A CN117874347A (en) | 2024-01-15 | 2024-01-15 | Content recommendation technology based on business characteristics |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117874347A true CN117874347A (en) | 2024-04-12 |
Family
ID=90591647
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410056798.5A Pending CN117874347A (en) | 2024-01-15 | 2024-01-15 | Content recommendation technology based on business characteristics |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117874347A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118427371A (en) * | 2024-07-02 | 2024-08-02 | 宁波瑞能智慧科技股份有限公司 | Multimedia presentation method and system based on data identification |
| CN118445493A (en) * | 2024-07-08 | 2024-08-06 | 杭州字节方舟科技有限公司 | Block chain big data recommendation method based on artificial intelligence |
| CN119046524A (en) * | 2024-07-01 | 2024-11-29 | 广东宏胜建材科技有限公司 | Intelligent customer demand analysis and material recommendation system |
-
2024
- 2024-01-15 CN CN202410056798.5A patent/CN117874347A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN119046524A (en) * | 2024-07-01 | 2024-11-29 | 广东宏胜建材科技有限公司 | Intelligent customer demand analysis and material recommendation system |
| CN118427371A (en) * | 2024-07-02 | 2024-08-02 | 宁波瑞能智慧科技股份有限公司 | Multimedia presentation method and system based on data identification |
| CN118427371B (en) * | 2024-07-02 | 2024-08-30 | 宁波瑞能智慧科技股份有限公司 | Multimedia presentation method and system based on data identification |
| CN118445493A (en) * | 2024-07-08 | 2024-08-06 | 杭州字节方舟科技有限公司 | Block chain big data recommendation method based on artificial intelligence |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117874347A (en) | Content recommendation technology based on business characteristics | |
| CN105426528B (en) | A kind of retrieval ordering method and system of commodity data | |
| Li | Harnessing multimodal data and mult-recall strategies for enhanced product recommendation in e-commerce | |
| CN103731738A (en) | Video recommendation method and device based on user group behavioral analysis | |
| CN105787061A (en) | Information pushing method | |
| CN106649658B (en) | Recommendation system and method for indiscriminate treatment of user roles and data sparse | |
| CN106897914A (en) | A kind of Method of Commodity Recommendation and system based on topic model | |
| Stanescu et al. | A hybrid recommender system: User profiling from keywords and ratings | |
| CN110688565B (en) | Next item recommendation method based on multidimensional Hawkes process and attention mechanism | |
| CN111651678A (en) | Knowledge graph-based personalized recommendation method | |
| CN116541607A (en) | Intelligent recommendation method based on commodity retrieval data analysis | |
| CN114861079A (en) | Collaborative filtering recommendation method and system fusing commodity features | |
| CN113705217A (en) | Literature recommendation method and device for knowledge learning in power field | |
| CN118014652A (en) | Advertisement creative design method and system based on artificial intelligence and big data analysis technology | |
| CN112115384A (en) | A Time Series Recommendation Algorithm Based on Generative Sorting | |
| CN104572915A (en) | User event relevance calculation method based on content environment enhancement | |
| She et al. | Research on advertising click-through rate prediction based on CNN-FM hybrid model | |
| CN106991592B (en) | A personalized recommendation method based on the analysis of purchasing user behavior | |
| Molina et al. | Recommendation system for netflix | |
| CN110727867A (en) | Semantic entity recommendation method based on fuzzy mechanism | |
| Agagu et al. | Context-aware recommendation methods | |
| Luo | Research on collaborative filtering of food information security in E-commerce platform | |
| CN112989196A (en) | Book Recommendation Method Based on Personalized Recall Algorithm LFM | |
| Chen et al. | Study on collaborative filtering recommendation algorithm based on web user clustering | |
| Wang et al. | Improved metric factorization recommendation algorithm based on social networks and implicit feedback |
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 |