CN113076472B - Movie recommendation method and system based on user requirements and label association degree - Google Patents
Movie recommendation method and system based on user requirements and label association degree Download PDFInfo
- Publication number
- CN113076472B CN113076472B CN202110280021.3A CN202110280021A CN113076472B CN 113076472 B CN113076472 B CN 113076472B CN 202110280021 A CN202110280021 A CN 202110280021A CN 113076472 B CN113076472 B CN 113076472B
- Authority
- CN
- China
- Prior art keywords
- items
- target
- movie
- item
- movies
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
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
- G06F16/40—Information retrieval; Database structures therefor; File system structures therefor of multimedia data, e.g. slideshows comprising image and additional audio data
- G06F16/43—Querying
- G06F16/435—Filtering based on additional data, e.g. user or group profiles
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/40—Information retrieval; Database structures therefor; File system structures therefor of multimedia data, e.g. slideshows comprising image and additional audio data
- G06F16/48—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Library & Information Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
A movie recommendation method based on user requirements and tag association degrees comprises the following steps: acquiring movie labels input by a user as target labels, and combining all the target labels to form a target item; acquiring all movie labels except the target label contained in the target item and the movie in the movie library as original items, and acquiring the original items with the support degree greater than or equal to the minimum support degree a0 as frequent items; combining the frequent items pairwise to serve as candidate items, and marking the candidate items with the support degree greater than or equal to the minimum support degree a0 as transition items; updating the frequent items into transition items and combining the frequent items two by two again; until the candidate item containing the target item is screened as a preparation item; the preparation item is divided into a target item and an associated item, and the associated item contains all movie labels except the target item in the preparation item. And then recommending the movies to the user according to the associated items, thereby ensuring the association between the recommended movies and the interest tendency of the user and realizing the accurate orientation of the movies recommended by the user.
Description
Technical Field
The invention relates to the field of data mining and information processing, in particular to a movie recommendation method and system based on user requirements and tag association degrees.
Background
With the development of internet technology and the increasing demand of people, and in recent years, data mining is widely used in various industries, such as movie recommendation for users. The existing movie recommendation algorithm mainly adopts a collaborative filtering algorithm, and the function of the algorithm is mainly to find users similar to the algorithm according to own hobbies and then recommend the hobbies of the similar users.
Collaborative filtering algorithms are based on the idea that item attributes or user scores are close. The main viewpoint is the selection and recommendation of users with similar viewpoints. Is an information filtering algorithm that relies on a large amount of user information. And searching a group of user sets and item sets with scores similar to the scores of the target users from a large number of users, predicting the scores of the target users for the items according to the scores of the similar users for the same item, and obtaining recommendation opinions to other users in the cluster according to the ranking sequence of the scores obtained by the items. At present, the collaborative filtering recommendation method has the following technical problems:
1. the recommendation algorithm needs to rely on the scoring information of other users for judgment, so that the information of other users needs to be collected and processed, and the recommendation efficiency is reduced.
2. The accuracy of recommendation by the recommendation algorithm is still problematic, and the recommendation algorithm recommends based on the scores of others so that the recommended movie may not be the type that the recommendation algorithm wants to see by itself, and therefore the accuracy is not high.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides a movie recommendation method and system based on user requirements and tag association, and aims to improve recommendation efficiency, improve recommendation relevance and perform relevant recommendation according to the requirements of users.
The invention adopts the following technical scheme:
a movie recommendation method based on user requirements and label association degrees comprises the following steps:
s1, acquiring a movie label input by a user as a target label, and combining all the target labels to form a target item; setting a minimum support degree a0 and a minimum confidence b0, wherein 0< a0 is not more than c0, and c0 is the value of the support degree of the target item; 0< b0< 1;
the calculation method of the support degree c is as follows: f is the number of the movies comprising the movie label set with the support degree to be calculated in the movie library, and the movie label set at least comprises one movie label; d is the total number of the movies in the movie library;
s2, acquiring all movie labels and target items except the target label contained in the movies in the movie library as original items, and acquiring the original items with the support degree greater than or equal to the minimum support degree a0 as frequent items;
s3, merging the frequent items pairwise to be candidate items, and marking the candidate items with the support degree greater than or equal to the minimum support degree a0 as transition items;
s4, judging whether the number of transition items is greater than or equal to 2, and at least one transition item comprises a target item; if yes, updating the frequent item into a transition item, and then returning to the step S3;
s5, if not, screening the candidate item containing the target item as a preparation item; the preparation items are divided into target items and associated items, and the associated items comprise all movie labels except the target items in the preparation items;
s6, calculating the confidence of each prepared item;
the confidence coefficient is calculated in the following way:
wherein the content of the first and second substances,representing a set of movie labels, y representing a target item, z representingA set of movie labels other than the medium target item,to representThe degree of confidence of (a) is,representSupport of (2)Degree, support (y) represents the support degree of the target item;
and S7, acquiring the association item of the preparation item with the confidence degree greater than the minimum confidence degree b0 as the target association item.
Preferably, step S6 specifically includes: calculating the confidence coefficient of each prepared item, and judging whether the prepared item with the confidence coefficient larger than or equal to the minimum confidence coefficient b0 exists; if yes, go to step S7; otherwise, go to step S8;
s8: and extracting frequent items containing the target items from the frequent items forming each prepared item as target supplementary items, and acquiring associated items of the target supplementary items with confidence degrees higher than the minimum confidence degree b0 as target associated items, wherein the associated items of the target supplementary items are combinations of movie labels except the target items in the target supplementary items.
Preferably, the method further comprises step S9: and screening the movies based on the target associated items and recommending the movies to the user.
Preferably, step S9 specifically includes: respectively acquiring corresponding recommended movies corresponding to the target association items, and sequencing the recommended movies according to the gradient of the target association items; the gradient of each target associated item is positively correlated with the confidence of the preparation item where the target associated item is located.
Preferably, the method for sorting the recommended movies includes:
the first bit: the highest-scoring movie among the movies containing the target association of the first gradient in the movie label;
second position: the highest-scoring movie among the movies containing the target association of the second gradient in the movie label;
……
an Nth position: the highest-scoring movie among the movies containing the target association of the nth gradient in the movie label;
position N + 1: the second highest scoring movie among the movies containing the target association of the first gradient in the movie label;
position N + 1: the movie with the second highest score among the movies with the target association in the movie label comprising the second gradient;
repeating the steps until the recommended position is filled;
n represents the number of target associated items, the target associated items of the first gradient to the target associated items of the Nth gradient, and the corresponding confidence degrees are gradually reduced.
Preferably, step S0 is further provided before step S1: setting a plurality of support degree thresholds and a plurality of confidence degree thresholds;
in step S1, a minimum support a0 is selected from the support thresholds smaller than c 0;
in step S7, from less than CON MAX Selecting the minimum confidence b0 from the confidence thresholds;
taking the original items with the support degree greater than or equal to the minimum support degree a0 as original frequent items, combining the original frequent items pairwise to serve as original candidate items, and taking the item with the maximum support degree in the original candidate items containing the target items as an original target item, wherein CON (CON) MAX Is the confidence of the original target item.
Preferably, in step S1, the support threshold closest to c0 and smaller than c0 is selected as the minimum support a 0.
Preferably, in step S7, the closest CON is selected M And is less than CON MAX As the minimum confidence b 0; CON M Is the maximum of the confidences of the preparatory terms.
A movie recommendation system based on user demand and tag association, comprising: the system comprises a storage module and a processor, wherein the storage module is used for storing a computer program, and the processor is used for realizing the movie recommendation method based on the user requirements and the label association degree when executing the computer program.
The invention has the advantages that:
(1) in the invention, the frequent item is taken from the movie label to obtain the target associated item closely related to the target label input by the user, and then the movie is recommended to the user according to the associated item, thereby ensuring the association between the recommended movie and the interest tendency of the user, realizing the accurate orientation of the movie recommended to the user and improving the user experience.
(2) The invention only needs to recommend according to the movie label, and has less number of labels of the movie compared with all information of other users, thereby better collecting, higher processing efficiency and higher recommending speed.
(3) Through the setting of the minimum support degree a0, the target item is guaranteed to have at least one chance of being merged with other movie labels.
(4) Through the establishment of the target item, the frequent items are merged based on the target item, so that the close correlation between the finally obtained preparation item and the user requirement is ensured, and meanwhile, the frequent items are prevented from being interfered by incomplete target labels aiming at the target item consisting of a plurality of movie labels, so that the screening efficiency of the frequent items is facilitated, and the working efficiency of the whole system is improved.
(5) And screening the prepared items according to the confidence degrees, thereby further ensuring the close association degree of the finally obtained target associated items and the target items. And the number of the finally obtained target associated items is controlled, so that accurate recommendation of the user is ensured.
(6)CON MAX The setting of the method ensures the effective acquisition of the target association and avoids the condition that the target association is an empty set; CON M The setting of (3) further ensures the close association degree of the target associated item and the target item, and ensures the high-precision orientation of recommending the film to the user.
Drawings
FIG. 1 is a flow chart of a movie recommendation method based on user demand and tag association;
fig. 2 is a flowchart of another movie recommendation method based on user requirements and tag association.
Detailed Description
The movie recommendation method based on the user requirements and the label association degree provided by the embodiment comprises the following steps.
S1, acquiring movie labels input by a user as target labels, and combining all the target labels to form target items; setting a minimum support degree a0 and a minimum confidence degree b0, wherein 0< a0 is not more than c0, and c0 is the value of the support degree of the target item; 0< b0< 1.
The calculation method of the support degree c is as follows: f is the number of the movies comprising the movie label set with the support degree to be calculated in the movie library, and the movie label set at least comprises one movie label; d is the total number of movies in the movie library.
Namely: the support degree of a single movie label is the ratio of the number of movies containing the movie label to D; the support of a tag set comprising a plurality of movie tags is the ratio of the number of movies comprising the tag set to D.
S2, all movie labels and target items except the target label contained in the movies in the movie library are obtained as original items, and the original items with the support degree larger than or equal to the minimum support degree a0 are obtained as frequent items.
And S3, merging the frequent items pairwise to serve as candidates, and marking the candidates with the support degree greater than or equal to the minimum support degree a0 as transition items.
S4, judging whether the number of transition items is greater than or equal to 2, and at least one transition item comprises a target item; if yes, the frequent item is updated to the transition item, and then the step S3 is returned to.
S5, if not, screening the candidate item containing the target item as a preparation item; the preparation item is divided into a target item and an associated item, and the associated item contains all movie labels except the target item in the preparation item, namely the associated item + the target item is the preparation item.
In this way, in the present embodiment, the setting of the minimum support degree a0 ensures that the target item has at least one chance to be merged with other movie labels. In this embodiment, the loop merging of the frequent items is realized by looping the steps S3 and S4, so that the movie labels closely associated with the target item are accurately searched, and the movie range required by the user is accurately oriented.
In addition, in the embodiment, through establishment of the target item, frequent items are merged based on the target item, so that the finally obtained preparation item is ensured to be closely related to the user requirement, and meanwhile, for the target item consisting of a plurality of movie labels, the frequent items are prevented from being interfered incompletely by the target labels, so that the screening efficiency of the frequent items is facilitated, and the working efficiency of the whole system is improved.
S6, calculating the confidence of each prepared item, and judging whether the prepared item with the confidence greater than the minimum confidence b0 exists.
The confidence coefficient is calculated in the following way:
wherein the content of the first and second substances,representing a set of movie labels, y representing a target item, z representingA set of movie labels other than the medium target item,to representThe degree of confidence of (a) is,to representSupport (y) represents the support of the target item.
And S7, if the preparation item with the confidence coefficient greater than or equal to the minimum confidence coefficient b0 exists, acquiring the association item of the preparation item with the confidence coefficient greater than the minimum confidence coefficient b0 as the target association item.
And S8, if no preparation item with the confidence coefficient greater than or equal to the minimum confidence coefficient b0 exists, extracting a frequent item containing the target item from the frequent items forming each preparation item as a target supplementary item, and acquiring an associated item of the target supplementary item with the confidence coefficient greater than the minimum confidence coefficient b0 as a target associated item, wherein the associated item of the target supplementary item is a target supplementary item-target item, namely the associated item of the target supplementary item is a combination of movie labels other than the target item in the target supplementary item.
And S9, screening the movies based on the target associated items and recommending the movies to the user.
In this way, in the present embodiment, the preliminary item is screened according to the confidence, so that the close association degree between the finally obtained target association item and the target item is further ensured. And the number of the finally obtained target associated items is controlled, so that accurate recommendation of the user is ensured.
Step S9 specifically includes: respectively acquiring corresponding recommended movies corresponding to the target association items, and sequencing the recommended movies according to the gradient of the target association items; the gradient of each target associated item is positively correlated with the confidence of the preparation item where the target associated item is located. Therefore, the movies corresponding to the target associated items with higher association degree of the target items are ensured to occupy more displayed recommendation positions, and the user experience is improved.
In this embodiment, the method of sorting the recommended movies includes:
the first bit: the highest-scoring movie among the movies containing the target association of the first gradient in the movie label;
second position: the highest-scoring movie among the movies containing the target association of the second gradient in the movie label;
……
the Nth position: the highest-scoring movie among the movies containing the target association of the nth gradient in the movie label;
position N + 1: the second highest scoring movie among the movies containing the target association of the first gradient in the movie label;
position N + 1: the movie with the second highest score among the movies with the target association in the movie label comprising the second gradient;
repeating the steps until the recommended position is filled;
n represents the number of target associated items, the target associated items of the first gradient to the target associated items of the Nth gradient, and the corresponding confidence degrees are gradually reduced.
Therefore, the recommended movies are screened according to the movie scores, and the quality of the recommended movies is guaranteed; through the cross arrangement of the movies under different target associated items, the richness of recommended movie types is ensured.
In specific implementation, step S0 may be provided before step S1 in this embodiment: a plurality of support degree thresholds and a plurality of confidence degree thresholds are set. As such, in step S1, the minimum support degree a0 may be selected from the support degree thresholds less than c0, for example, the support degree threshold closest to c0 and less than c0 is selected as the minimum support degree a 0.
In step S7, from less than CON MAX Selecting a minimum confidence b0 from the confidence threshold values, wherein the original items with the support degree greater than or equal to the minimum support degree a0 are used as original frequent items, the original frequent items are combined pairwise to be used as original candidate items, the item with the maximum support degree in the original candidate items containing the target items is used as an original target item, and the CON MAX Is the confidence of the original target item. CON MAX The setting of (2) ensures the effective acquisition of the target association item, and avoids the condition that the target association item is an empty set. In particular, the closest CON may be selected M And is less than CON MAX As the minimum confidence b0, CON M And the maximum value of the confidence degrees of the prepared items is used for ensuring the close association degree of the target associated item and the target item and ensuring the high-precision orientation of recommending the movie to the user.
The embodiment further provides a movie recommendation system based on user requirements and tag association degrees, which includes: the system comprises a storage module and a processor, wherein the storage module is used for storing a computer program, and the processor is used for realizing the movie recommendation method based on the user requirements and the label association degree when executing the computer program.
The above method is further explained below with reference to a specific example.
In the following examples, assuming that a movie library contains 10 movies, the movie labels are as follows:
movie 1: classical philosophy of science fiction us american drama
Movie 2: super hero environment-friendly warmer for recomposing Chinese cartoon with science fiction love plot
Movie 3: science fiction us super hero cartoon recomposition romance
Movie 4: super hero action of science fiction us drama
Movie 5: adaptation of science fiction us super hero caricature
Movie 6: suspicion of disaster, U.S. terror
Movie 7: magic history American war classic
Movie 8: science fiction growth friendship american cartoon
Movie 9: japanese science fiction classical special pickup
Movie 10: thriller and thrill on victory
Example 1
Assume that the user's nail enters the wisdom label "science fiction", "us", and thus the target item "science fiction, us" is available.
Movies 1, 4, 5, 8 all contain the label "science fiction, usa" at the same time, so the target item has a support of 40%. Thus, in the present embodiment, the minimum support degree a0 may be set to 30%.
Referring to table 1, in the present embodiment, according to the minimum support degree, the frequent items shown in table 1 may be screened from the original items:
table 1: first time frequent item statistical table
Table 2: candidate statistic table composed of frequent items in table 1
From table 2, the frequent items "usa, science fiction, super hero" and "super hero, caricature adaptation" can be screened, and "usa, science fiction, super hero" includes the target item. Therefore, the 'American, science fiction, super hero' and 'super hero, cartoon adaptation' can be merged again to obtain candidate items 'American, science fiction, super hero, cartoon adaptation'. In the merging process, only one candidate item is obtained, so that merging can not be carried out again.
"American, science fiction, super hero, caricature" was adapted as a preliminary item.
The support degree of the preparation item is 20%;
the confidence of the preliminary term is 20%/50%/40%.
CON M =20%/50%=40%。
therefore, 40% b0 is less than or equal to 60%.
Assuming that the minimum confidence b0 is set to 40% in the present embodiment, the target association is obtained as "super hero, caricature adaptation".
In this embodiment, the screenable movie tag contains a high-quality movie of "super hero, comic adaptation" and is recommended to the user a.
Example 2
In this embodiment, the minimum confidence b0 is 50% as compared with embodiment 1.
Thus, in the present embodiment, the confidence of the preliminary item is 40% < b0, so that a frequent set of "american, science fiction, super hero, comic adaptation" constituting the preliminary item is obtained, specifically referring to table 2:
"american, science fiction, super hero, caricature recomposition" + "super hero, caricature recomposition"
Thus, the target supplement item "American, science fiction, super hero" is available.
The confidence of this "american, science fiction, super hero" is 30%/50% > 60% > b 0.
Therefore, in this embodiment, the obtained target association is "super hero"; then, the high-quality movies whose movie labels contain "super hero" can be screened and recommended to the user A.
The present invention is not limited to the above embodiments, and any modifications, equivalent substitutions and improvements made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (8)
1. A movie recommendation method based on user requirements and label association degrees is characterized by comprising the following steps:
s1, acquiring movie labels input by a user as target labels, and combining all the target labels to form target items; setting a minimum support degree a0 and a minimum confidence degree b0, wherein 0< a0 is not more than c0, and c0 is the value of the support degree of the target item; 0< b0< 1;
the calculation method of the support degree c is as follows: f is the number of the movies of the movie label set containing the supportability to be calculated in the movie library, and the movie label set at least contains one movie label; d is the total number of the movies in the movie library;
s2, acquiring all movie labels and target items except the target label contained in the movies in the movie library as original items, and acquiring the original items with the support degree greater than or equal to the minimum support degree a0 as frequent items;
s3, merging every two frequent items as candidate items, and recording the candidate items with the support degree greater than or equal to the minimum support degree a0 as transition items;
s4, judging whether the number of transition items is greater than or equal to 2, and at least one transition item comprises a target item; if yes, the frequent item is updated to the transition item, and then the step S3 is returned;
s5, if not, screening the candidate item containing the target item as a preparation item; the preparation items are divided into target items and associated items, and the associated items comprise all movie labels except the target items in the preparation items;
s6, calculating the confidence of each prepared item;
the confidence coefficient is calculated in the following way:
wherein the content of the first and second substances,representing a set of movie labels, y representing a target item, z representingA set of movie labels other than the medium target item,to representThe degree of confidence of (a) is,representSupport (y) represents the support of the target item;
s7, acquiring the associated item of the preparation item with the confidence coefficient larger than the minimum confidence coefficient b0 as a target associated item;
and S9, screening the movies based on the target associated items and recommending the movies to the user.
2. The movie recommendation method based on user requirements and label association as claimed in claim 1, wherein step S6 specifically comprises: calculating the confidence coefficient of each prepared item, and judging whether the prepared item with the confidence coefficient greater than or equal to the minimum confidence coefficient b0 exists; if yes, go to step S7; otherwise, go to step S8;
s8: and extracting frequent items containing the target items from the frequent items forming each prepared item as target supplementary items, and acquiring associated items of the target supplementary items with confidence degrees higher than the minimum confidence degree b0 as target associated items, wherein the associated items of the target supplementary items are combinations of movie labels except the target items in the target supplementary items.
3. The movie recommendation method based on user requirements and label association as claimed in claim 1, wherein step S9 specifically comprises: respectively acquiring corresponding recommended movies corresponding to the target association items, and sequencing the recommended movies according to the gradient of the target association items; the gradient of each target associated item is positively correlated with the confidence of the preparation item where the target associated item is located.
4. The movie recommendation method based on user demand and tag association as claimed in claim 3, wherein the recommended movies are ranked in a manner of:
the first bit: the highest-scoring movie of the movies containing the target association of the first gradient in the movie label;
second position: the highest-scoring movie of the movies containing the target association of the second gradient in the movie label;
……
the Nth position: the highest-scoring movie among the movies containing the target association of the nth gradient in the movie label;
position N + 1: the second highest scoring movie among the movies containing the target association of the first gradient in the movie label;
position N + 1: the movie with the second highest score among the movies with the target association in the movie label comprising the second gradient;
repeating the steps until the recommended position is filled;
n represents the number of target associated items, the target associated items of the first gradient to the target associated items of the Nth gradient, and the corresponding confidence degrees are gradually reduced.
5. The movie recommendation method based on user requirements and tag association as claimed in claim 1, wherein step S1 is preceded by step S0: setting a plurality of support degree thresholds and a plurality of confidence degree thresholds;
in step S1, a minimum support a0 is selected from the support thresholds smaller than c 0;
in step S7, from less than CON MAX Selecting the minimum confidence b0 from the confidence thresholds;
wherein, the original items with the support degree greater than or equal to the minimum support degree a0 are used as original frequent items, and the original frequent items are combined pairwise to be used as original candidatesOptions, the item with the highest support among the original candidate items containing the target item being denoted as the original target item, CON MAX Is the confidence of the original target item.
6. The movie recommendation method based on user demand and tag association as claimed in claim 5, wherein in step S1, the support threshold closest to c0 and smaller than c0 is selected as the minimum support a 0.
7. The movie recommendation method based on user demand and tag association as claimed in claim 5, wherein in step S7, the closest CON is selected M And is less than CON MAX As the minimum confidence b 0; CON M Is the maximum of the confidences of the preparatory terms.
8. A movie recommendation system based on user demand and tag association, comprising: a storage module for storing a computer program, and a processor for implementing the movie recommendation method based on user requirements and tag association as claimed in any one of claims 1 to 7 when executing the computer program.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110280021.3A CN113076472B (en) | 2021-03-16 | 2021-03-16 | Movie recommendation method and system based on user requirements and label association degree |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110280021.3A CN113076472B (en) | 2021-03-16 | 2021-03-16 | Movie recommendation method and system based on user requirements and label association degree |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113076472A CN113076472A (en) | 2021-07-06 |
CN113076472B true CN113076472B (en) | 2022-09-16 |
Family
ID=76612524
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110280021.3A Active CN113076472B (en) | 2021-03-16 | 2021-03-16 | Movie recommendation method and system based on user requirements and label association degree |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113076472B (en) |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105719189A (en) * | 2016-01-15 | 2016-06-29 | 天津大学 | Tag recommendation method for effectively increasing tag diversity in social network |
CN110517449A (en) * | 2019-09-25 | 2019-11-29 | 合肥工业大学 | A kind of intelligent air control system for net about vehicle, taxi |
CN111191127A (en) * | 2019-12-24 | 2020-05-22 | 重庆特斯联智慧科技股份有限公司 | Travel recommendation method and system based on correlation analysis algorithm |
CN111339427A (en) * | 2020-03-23 | 2020-06-26 | 卓尔智联(武汉)研究院有限公司 | Book information recommendation method, device and system and storage medium |
CN111932342A (en) * | 2020-08-12 | 2020-11-13 | 中国银行股份有限公司 | User cold start product recommendation method and system based on Apriori algorithm |
CN112085618A (en) * | 2020-06-16 | 2020-12-15 | 西安交通大学 | Frequent item set-based power distribution equipment fault factor relevance mining method |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108763507A (en) * | 2018-05-30 | 2018-11-06 | 北京百度网讯科技有限公司 | Enterprise's incidence relation method for digging and device |
-
2021
- 2021-03-16 CN CN202110280021.3A patent/CN113076472B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105719189A (en) * | 2016-01-15 | 2016-06-29 | 天津大学 | Tag recommendation method for effectively increasing tag diversity in social network |
CN110517449A (en) * | 2019-09-25 | 2019-11-29 | 合肥工业大学 | A kind of intelligent air control system for net about vehicle, taxi |
CN111191127A (en) * | 2019-12-24 | 2020-05-22 | 重庆特斯联智慧科技股份有限公司 | Travel recommendation method and system based on correlation analysis algorithm |
CN111339427A (en) * | 2020-03-23 | 2020-06-26 | 卓尔智联(武汉)研究院有限公司 | Book information recommendation method, device and system and storage medium |
CN112085618A (en) * | 2020-06-16 | 2020-12-15 | 西安交通大学 | Frequent item set-based power distribution equipment fault factor relevance mining method |
CN111932342A (en) * | 2020-08-12 | 2020-11-13 | 中国银行股份有限公司 | User cold start product recommendation method and system based on Apriori algorithm |
Non-Patent Citations (2)
Title |
---|
Frequent Pattern Mining Based On Occupation and Correlation;Kai Zhang,等;《2020 IEEE 3rd International Conference on Electronic Information and Communication Technology (ICEICT)》;20210201;第161-166页 * |
基于规则的推荐算法分析和实现;段继光,等;《电子技术与软件工程》;20201130(第21期);第140-141页 * |
Also Published As
Publication number | Publication date |
---|---|
CN113076472A (en) | 2021-07-06 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US7502780B2 (en) | Information storage and retrieval | |
US9342614B2 (en) | Asymmetric identification of interest twins in an online community | |
EP1565846B1 (en) | Information storage and retrieval | |
CN101055585B (en) | System and method for clustering documents | |
EP2829990B1 (en) | Image search device, image search method, program, and computer-readable storage medium | |
US20130332441A1 (en) | Systems and Methods for Identifying Terms Relevant to Web Pages Using Social Network Messages | |
US7827172B2 (en) | “Query-log match” relevance features | |
EP1426882A2 (en) | Information storage and retrieval | |
US20090171942A1 (en) | Predicting and ranking search query results | |
US8732116B1 (en) | Harvesting relational tables from lists on the web | |
US20140344303A1 (en) | Method, system and controller for searching a database containing data items | |
CN105589929A (en) | Image retrieval method and device | |
CN104077385A (en) | Classification and retrieval method of files | |
JP2008084203A (en) | System, method and program for assigning label | |
CN111428120B (en) | Information determination method and device, electronic equipment and storage medium | |
CN113076472B (en) | Movie recommendation method and system based on user requirements and label association degree | |
CN111753043B (en) | Document data processing method, device and storage medium | |
CN111737461B (en) | Text processing method and device, electronic equipment and computer readable storage medium | |
CN103186650A (en) | Searching method and device | |
US10242080B1 (en) | Clustering applications using visual metadata | |
CN111190893A (en) | Method and device for establishing feature index | |
CN107169065B (en) | Method and device for removing specific content | |
CN106572394B (en) | Movie and television data navigation method | |
Kang et al. | Recognising informative Web page blocks using visual segmentation for efficient information extraction. | |
CN108427759A (en) | Real time data computational methods for mass data processing |
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 |