CN113806612A - Method for detecting key community in user movie network based on index - Google Patents
Method for detecting key community in user movie network based on index Download PDFInfo
- Publication number
- CN113806612A CN113806612A CN202111142284.4A CN202111142284A CN113806612A CN 113806612 A CN113806612 A CN 113806612A CN 202111142284 A CN202111142284 A CN 202111142284A CN 113806612 A CN113806612 A CN 113806612A
- Authority
- CN
- China
- Prior art keywords
- community
- user
- compression
- movie
- index
- 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
- 238000000034 method Methods 0.000 title claims abstract description 25
- 238000010276 construction Methods 0.000 claims abstract description 13
- 230000006835 compression Effects 0.000 claims description 102
- 238000007906 compression Methods 0.000 claims description 102
- 230000000694 effects Effects 0.000 claims description 7
- 238000001514 detection method Methods 0.000 abstract description 7
- 238000002474 experimental method Methods 0.000 description 4
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000003012 network analysis Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
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
- G06F16/951—Indexing; Web crawling techniques
-
- 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
- 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
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- 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
- G06Q50/01—Social networking
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Development Economics (AREA)
- Marketing (AREA)
- Educational Administration (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a method for detecting a key community in a user movie network based on indexes. In order to find a key community which simultaneously meets the constraint of the number of film watching parts of a user and the constraint of film scoring, the invention provides a new key community model on a user film network, namely a (k, s) -community model, which meets three conditions: each user of the user layer watches at least k movies; the total score of each movie of the movie layer is not less than s, wherein the total score is the sum of the scores of all connected users; communities are extremely large, i.e., there is no community of any size greater than it meets the first two conditions. In consideration of the computational complexity of the model, the invention provides an efficient index-based detection algorithm, so that the (k, s) -community model can be quickly found in a large-scale user movie network. Meanwhile, the index structure can reduce the index construction cost and realize the balance between space consumption and query efficiency.
Description
Technical Field
The invention belongs to the technical field of multimedia data detection, and particularly relates to a method for detecting a key community in a user movie network based on indexes.
Background
With the increasingly rich cultural life of people, movies begin to become inseparable from people's recreational life. There are many interest groups in the social network of today, and the movie preference group is also an important component of them. Therefore, in the social network, the research of a network model such as a user movie is of great significance for discovering interest key communities in the social network. However, the user movie network has a certain specificity, and they are composed of two sets of mutually disjoint sets, namely, a user set and a movie set, so that they cannot be represented by a simple network structure. Therefore, we model the user movie network as a bipartite graph G ═ U, L, E, consisting of two sets of mutually disjoint upper user set U and lower movie set L, where an edge can only connect users and movies from different sets, i.e. two end points of an edge in the edge set E can only exist in one user set U and the other in the movie set L.
As a fundamental problem in network analysis, the detection of key communities has been extensively studied. However, the general model only focuses on the cohesion of the relationship connection between entities, and neglects another important feature, namely the interaction weight between different entities. In real life, the edges of a user movie network usually have a certain weight, that is, the score of the user for the movie reflects the interest degree of the user for the movie, and is an important index for detecting a key community in the social network. Therefore, ignoring this feature would leave us missing a very useful piece of information.
Disclosure of Invention
In order to find a key community which simultaneously meets the constraint of the number of movie watching parts of a user and the constraint of rating, the invention provides a new key community model on a movie network of the user, namely a (k, s) -community model, which meets three conditions: each user of the user layer watches at least k movies; the total score of each movie of the movie layer is not less than s, wherein the total score is the sum of the scores of all connected users; communities are extremely large, i.e., there is no community of any size greater than it meets the first two conditions.
The invention develops an efficient index-based detection algorithm by utilizing the nesting characteristic of the (k, s) -community model, and can quickly find the (k, s) -community model in a large-scale user movie network. In addition, the index structure provided by the invention can reduce the construction cost of the index and achieve the balance between space consumption and query efficiency.
The technical scheme for solving the technical problems of the invention is as follows: a method for detecting key communities in a user movie network based on indexes, wherein the key communities are in a (k, s) -community model and satisfy three conditions: each user of the user layer watches at least k movies; the total score of each movie of the movie layer is not less than s, wherein the total score is the sum of the scores of all connected users; communities are extremely large, i.e., there is no community of any size greater than it meets the first two conditions; the method comprises the following steps:
step one, according to the nesting characteristic of a (k, s) -community, index construction is carried out on a user movie network G, and the index construction comprises the following steps:
(a) creating an index table and a compression table, wherein the index table and the compression table respectively store the residual user movie objects and compression directions after compression by taking k as a row sequence number and s as a column sequence number, and the compression comprises row compression in the direction of s and column compression in the direction of k; initializing the index table and compression table to null;
(b) starting iterative computation from k equal to 0;
(c) initializing s to be 0;
(d) removing users which do not meet the requirement that the number of watching parts is more than or equal to k +1 times from the (k, s) -community to obtain the (k +1, s) -community, and updating, namely updating the total score of each movie connected by the removal user to the original total score minus the score of the removal user for the movie; similarly, removing the movies which do not meet the total score and are greater than or equal to s +1 to obtain a (k, s +1) -community, and updating, namely, removing the number of watching parts of each user connected with the movies and updating to reduce the number of original parts by one;
(e) if the (k +1, s) -community is found to be contained in the (k, s +1) -community, the (k, s +1) -community is larger, the compression of the (k, s) -community to the (k, s +1) -community can reduce the number of stored users and the number of movies, and the line compression effect is better; therefore, the index table item [ k ] [ s ] stores the user movie set left after compressing (k, s) -community to (k, s +1) -community, and the compression table item [ k ] [ s ] stores the compression direction "→";
(f) if the (k, s +1) -community is found to be contained in the (k +1, s) -community, which indicates that the (k +1, s) -community is larger, the compression of the (k, s) -community to the (k +1, s) -community can reduce the number of stored users and the number of movies, i.e. the column compression effect is better; therefore, the index table item [ k ] [ s ] stores the user movie set remained after compressing (k, s) -to (k +1, s) -communities, and the compression table item [ k ] [ s ] stores the compression direction "↓";
(g) if both (e) and (f) are not true, then performing row compression and column compression simultaneously; the index table item [ k ] [ s ] stores the user movie sets left after compression from (k, s) -community to (k, s +1) -community and (k +1, s) -community, and the compression table item [ k ] [ s ] stores the compression directions "→" and "↓";
(h) adding 1 to the value of s and then repeating steps (d) to (h) until s has a maximum value;
(i) adding 1 to the value of k and then repeating steps (c) to (i) until k has a maximum value;
accessing the index table according to the given k, s value, adding the users and movies recorded in the index table item [ k ] [ s ] into the (k, s) -community user movie set, accessing the next index table item according to the compression direction recorded in the compression table item [ k ] [ s ], and adding the users and movies recorded in the next index table item into the (k, s) -community user movie set; the index table and the compression table are accessed alternately in an iteration mode, and the iteration is stopped until the compression direction in the accessed compression table entry is null;
and step three, outputting the sub-communities formed by all the users and the movies in the (k, s) -community user movie set, namely the (k, s) -communities to be detected.
Further, the nesting characteristic of the (k, s) -community is specifically: given a user movie network G, if there are k '≧ k and s' ≧ s, then the (k ', s') -community is nested in the (k, s) -community.
Further, the second step comprises:
(a) initializing a (k, s) -community user movie set to an empty set;
(b) finding out a corresponding index item [ k ] [ s ] in the index table according to the k, s values, and then adding the users and the movies recorded in the index item [ k ] [ s ] into a (k, s) -community user movie set;
(c) obtaining a corresponding compression direction in the compression table according to the k, s values;
(d) if the direction is only "→", then next find the index entry [ k ] [ s +1] and add the user and movie recorded therein to the (k, s) -community user movie collection;
(e) if the direction is only "↓", then finding an index table item [ k +1] [ s ] and adding the user and the movie recorded in the index table item into the (k, s) -community user movie set;
(f) if two directions are "→" and "↓", then finding an index table item [ k ] [ s +1] and an index table item [ k +1] [ s ], adding the recorded user and movie in the index table item [ k ] [ s ] to the (k, s) -community user movie set and ensuring that the user and the movie do not repeatedly appear in the user movie set;
(g) and (4) updating the k and s values according to the index table entry newly found at present, repeating the steps (c) to (g) until the compression direction in the accessed compression table entry is null, and stopping iteration.
The invention also discloses a server, which comprises a processor and a memory, wherein at least one instruction, at least one program, a code set or an instruction set is stored in the memory, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by the processor to realize the steps in the method for detecting the key communities in the user movie network based on the index.
The invention also discloses a computer readable storage medium, wherein at least one instruction, at least one program, code set or instruction set is stored in the storage medium, and the at least one instruction, the at least one program, the code set or instruction set is loaded and executed by a processor to realize the steps in the method for detecting the key communities in the user movie network based on the index.
The invention has the beneficial effects that: in order to obtain more effective user community information, the invention provides a novel (k, s) -community model, namely users in a key community meet certain requirements on the number of watched movies and each movie has a certain amount of scores. In consideration of the complexity of large-scale network processing and query, the invention provides an index-based detection algorithm, which can quickly find a (k, s) -community model in a large-scale user movie network and can reduce the space cost overhead of index construction. Therefore, the application of the method for detecting the key communities in the user movie network based on the index has great benefits for mining high-quality movie interest groups in the social network and helping the user to perform movie recommendation and movie evaluation analysis.
Drawings
FIG. 1 is a flow chart of a method for detecting key communities in a movie network of users based on indexing according to the present invention;
FIG. 2 is a schematic diagram of an original user movie network;
FIG. 3 is an index representation intent of an original user movie network;
fig. 4 is a compressed representation of the original user movie network.
Detailed Description
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in detail below.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, but the present invention may be practiced in other ways than those specifically described and will be readily apparent to those of ordinary skill in the art without departing from the spirit of the present invention, and therefore the present invention is not limited to the specific embodiments disclosed below.
The method for detecting the key community in the user movie network based on the index comprises a smart index construction strategy and an efficient index-based (k, s) -community detection algorithm. The implementation of each part is described in detail below.
The index construction strategy is to construct an index by using the nesting characteristic of a (k, s) -community model, and the characteristic can be used for helping to reduce the index construction cost, and specifically includes the following contents:
And (3) proving that: if a sub-community exists in the user movie network as (k ', s') -community, and k '≧ k, s' ≧ s. Then, according to the definition of the model, because k '≧ k and s' ≧ s, the (k ', s') -community is either a non-very large (k, s) -community or a very large (k, s) -community, contained within a very large (k, s) -community. Thus, lemma 1 is correct.
The efficient index-based (k, s) -community detection algorithm, as shown in fig. 1, specifically includes the following steps:
step one, index construction is carried out on a user movie network G according to the theorem 1, and the index construction comprises the following steps:
(a) creating an index table and a compression table, wherein the index table and the compression table respectively store the residual user movie objects and compression directions after compression by taking k as a row sequence number and s as a column sequence number, and the compression comprises row compression in the direction of s and column compression in the direction of k; initializing the index table and compression table to null;
(b) starting iterative computation from k equal to 0;
(c) initializing s to be 0;
(d) removing users which do not meet the requirement that the number of watching parts is more than or equal to k +1 times from the (k, s) -community to obtain the (k +1, s) -community, and updating, namely updating the total score of each movie connected by the removal user to the original total score minus the score of the removal user for the movie; similarly, removing the movies which do not meet the total score and are greater than or equal to s +1 to obtain a (k, s +1) -community, and updating, namely, removing the number of watching parts of each user connected with the movies and updating to reduce the number of original parts by one;
(e) if the (k +1, s) -community is found to be contained in the (k, s +1) -community, the (k, s +1) -community is larger, the compression of the (k, s) -community to the (k, s +1) -community can reduce the number of stored users and the number of movies, and the line compression effect is better; therefore, the index table item [ k ] [ s ] stores the user movie set left after compressing (k, s) -community to (k, s +1) -community, and the compression table item [ k ] [ s ] stores the compression direction "→";
(f) if the (k, s +1) -community is found to be contained in the (k +1, s) -community, which indicates that the (k +1, s) -community is larger, the compression of the (k, s) -community to the (k +1, s) -community can reduce the number of stored users and the number of movies, i.e. the column compression effect is better; therefore, the index table item [ k ] [ s ] stores the user movie set remained after compressing (k, s) -to (k +1, s) -communities, and the compression table item [ k ] [ s ] stores the compression direction "↓";
(g) if both (e) and (f) are not true, then performing row compression and column compression simultaneously; the index table item [ k ] [ s ] stores the user movie sets left after compression from (k, s) -community to (k, s +1) -community and (k +1, s) -community, and the compression table item [ k ] [ s ] stores the compression directions "→" and "↓";
(h) adding 1 to the value of s and then repeating steps (d) to (h) until s has a maximum value;
(i) adding 1 to the value of k and then repeating steps (c) to (i) until k has a maximum value;
accessing the index table according to the given k, s value, adding the users and movies recorded in the index table item [ k ] [ s ] into the (k, s) -community user movie set, accessing the next index table item according to the compression direction recorded in the compression table item [ k ] [ s ], and adding the users and movies recorded in the next index table item into the (k, s) -community user movie set; the index table and the compression table are accessed alternately in an iteration mode, and the iteration is stopped until the compression direction in the accessed compression table entry is null; the method specifically comprises the following substeps:
(a) initializing a (k, s) -community user movie set to an empty set;
(b) finding out a corresponding index item [ k ] [ s ] in the index table according to the k, s values, and then adding the users and the movies recorded in the index item [ k ] [ s ] into a (k, s) -community user movie set;
(c) obtaining a corresponding compression direction in the compression table according to the k, s values;
(d) if the direction is only "→", then next find the index entry [ k ] [ s +1] and add the user and movie recorded therein to the (k, s) -community user movie collection;
(e) if the direction is only "↓", then finding an index table item [ k +1] [ s ] and adding the user and the movie recorded in the index table item into the (k, s) -community user movie set;
(f) if two directions are "→" and "↓", then finding an index table item [ k ] [ s +1] and an index table item [ k +1] [ s ], adding the recorded user and movie in the index table item [ k ] [ s ] to the (k, s) -community user movie set and ensuring that the user and the movie do not repeatedly appear in the user movie set;
(g) and (4) updating the k and s values according to the index table entry newly found at present, repeating the steps (c) to (g) until the compression direction in the accessed compression table entry is null, and stopping iteration.
And step three, outputting the sub-communities formed by all the users and the movies in the (k, s) -community user movie set, namely the (k, s) -communities to be detected.
The effect of the present invention will be described below by taking the user movie network shown in fig. 2 as an example. FIG. 3 and FIG. 4 are respectively an index table and a compression table of FIG. 2, and the construction process thereof takes the generation of an index table item [1] [2] and a corresponding compression table item [1] [2] as an example: since the (1, 3) -community user movie set is { a1, a2, a3, a4, m1, m3, m4, m5}, (2, 2) -community user movie set is { a2, a3, a4, m2, m3, m4, m5}, the available (1, 3) -community does not contain the (2, 2) -community, and the (2, 2) -community does not contain the (1, 3) -community, it is necessary to perform row compression and column compression at the same time, and record compression directions "→, ↓" in the corresponding compression table item [1] [2 ]. Meanwhile, since the user movie set of the (1, 2) -community minus the union of the (1, 3) -community and the (2, 2) -community is empty, the corresponding index table entry [1] [2] is empty.
The following describes the query process of index-based (k, s) -communities, taking (2, 3) -communities as an example: initializing a (2, 3) -community user movie set as an empty set, firstly finding an index table item [2] [3], adding the recorded user and movie { a4, m5} into the (2, 3) -community set, wherein the (2, 3) -community user movie set is { a4, m5 }; then finding the corresponding compression table item [2] [3] to obtain a compression direction "→", finding the index table item [2] [4] according to the direction, wherein the recorded user and movie are empty, so that the (2, 3) -community user movie set is still { a4, m5 }; then finding the corresponding compression table item [2] [4] to obtain a compression direction "→", finding an index table item [2] [5] according to the direction, and adding the recorded user and movie { a2, a3, m3, m4} into the (2, 3) -community user movie set, wherein the (2, 3) -community user movie set is { a2, a3, a4, m3, m4, m5 }; and finally, finding out the corresponding compression table item [2] [5], finding out the compression table item is empty, and ending the iteration. Therefore, the sub-community formed by a2, a3, a4, m3, m4, m5 and its edges in the (2, 3) -community user movie set is the (2, 3) -community. It can be found that (2, 3) -communities composed of { a2, a3, a4, m3, m4, m5} have more close relationship between users and movies and the movies have higher scores, and are a high-quality movie interest group in the network, which can help in movie recommendation and movie evaluation analysis, compared with the original user movie network diagram shown in fig. 2.
In addition, the present invention performed extensive experiments on six networks to evaluate the effectiveness and efficiency of the proposed method. To evaluate the performance of the proposed method, we performed experiments by varying the parameters k and s. The invention measures the effectiveness of the proposed method with the algorithm time consumption and the index space cost. For each setting, the invention was run 200 times and averaged. All programs are implemented in standard c + +, and all experiments are performed on a PC equipped with an Intel Xeon 3.2GHz CPU and a 32GB RAM main memory. Experiments show that the index-based algorithm provided by the invention is 42 times faster than the basic algorithm.
The foregoing is only a preferred embodiment of the present invention, and although the present invention has been disclosed in the preferred embodiments, it is not intended to limit the present invention. Those skilled in the art can make numerous possible variations and modifications to the present teachings, or modify equivalent embodiments to equivalent variations, without departing from the scope of the present teachings, using the methods and techniques disclosed above. Therefore, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present invention are still within the scope of the protection of the technical solution of the present invention, unless the contents of the technical solution of the present invention are departed.
Claims (5)
1. A method for detecting key communities in a user movie network based on indexes is characterized in that the key communities are (k, s) -community models and satisfy three conditions: each user of the user layer watches at least k movies; the total score of each movie of the movie layer is not less than s, wherein the total score is the sum of the scores of all connected users; communities are extremely large, i.e., there is no community of any size greater than it meets the first two conditions; the method comprises the following steps:
step one, according to the nesting characteristic of a (k, s) -community, index construction is carried out on a user movie network G, and the index construction comprises the following steps:
(a) creating an index table and a compression table, wherein the index table and the compression table respectively store the residual user movie objects and compression directions after compression by taking k as a row sequence number and s as a column sequence number, and the compression comprises row compression in the direction of s and column compression in the direction of k; initializing the index table and compression table to null;
(b) starting iterative computation from k equal to 0;
(c) initializing s to be 0;
(d) removing users which do not meet the requirement that the number of watching parts is more than or equal to k +1 times from the (k, s) -community to obtain the (k +1, s) -community, and updating, namely updating the total score of each movie connected by the removal user to the original total score minus the score of the removal user for the movie; similarly, removing the movies which do not meet the total score and are greater than or equal to s +1 to obtain a (k, s +1) -community, and updating, namely, removing the number of watching parts of each user connected with the movies and updating to reduce the number of original parts by one;
(e) if the (k +1, s) -community is found to be contained in the (k, s +1) -community, the (k, s +1) -community is larger, the compression of the (k, s) -community to the (k, s +1) -community can reduce the number of stored users and the number of movies, and the line compression effect is better; therefore, the index table item [ k ] [ s ] stores the user movie set left after compressing (k, s) -community to (k, s +1) -community, and the compression table item [ k ] [ s ] stores the compression direction "→";
(f) if the (k, s +1) -community is found to be contained in the (k +1, s) -community, which indicates that the (k +1, s) -community is larger, the compression of the (k, s) -community to the (k +1, s) -community can reduce the number of stored users and the number of movies, i.e. the column compression effect is better; therefore, the index table item [ k ] [ s ] stores the user movie set remained after compressing (k, s) -to (k +1, s) -communities, and the compression table item [ k ] [ s ] stores the compression direction "↓";
(g) if both (e) and (f) are not true, then performing row compression and column compression simultaneously; the index table item [ k ] [ s ] stores the user movie sets left after compression from (k, s) -community to (k, s +1) -community and (k +1, s) -community, and the compression table item [ k ] [ s ] stores the compression directions "→" and "↓";
(h) adding 1 to the value of s and then repeating steps (d) to (h) until s has a maximum value;
(i) adding 1 to the value of k and then repeating steps (c) to (i) until k has a maximum value;
accessing the index table according to the given k, s value, adding the users and movies recorded in the index table item [ k ] [ s ] into the (k, s) -community user movie set, accessing the next index table item according to the compression direction recorded in the compression table item [ k ] [ s ], and adding the users and movies recorded in the next index table item into the (k, s) -community user movie set; the index table and the compression table are accessed alternately in an iteration mode, and the iteration is stopped until the compression direction in the accessed compression table entry is null;
and step three, outputting the sub-communities formed by all the users and the movies in the (k, s) -community user movie set, namely the (k, s) -communities to be detected.
2. The method for detecting key communities in a movie network of users based on indexing as claimed in claim 1, wherein the nesting characteristics of the (k, s) -communities are specifically: given a user movie network G, if there are k '≧ k and s' ≧ s, then the (k ', s') -community is nested in the (k, s) -community.
3. The method for detecting key communities in a movie network of users based on indexing as claimed in claim 1, wherein the second step comprises:
(a) initializing a (k, s) -community user movie set to an empty set;
(b) finding out a corresponding index item [ k ] [ s ] in the index table according to the k, s values, and then adding the users and the movies recorded in the index item [ k ] [ s ] into a (k, s) -community user movie set;
(c) obtaining a corresponding compression direction in the compression table according to the k, s values;
(d) if the direction is only "→", then next find the index entry [ k ] [ s +1] and add the user and movie recorded therein to the (k, s) -community user movie collection;
(e) if the direction is only "↓", then finding an index table item [ k +1] [ s ] and adding the user and the movie recorded in the index table item into the (k, s) -community user movie set;
(f) if two directions are "→" and "↓", then finding an index table item [ k ] [ s +1] and an index table item [ k +1] [ s ], adding the recorded user and movie in the index table item [ k ] [ s ] to the (k, s) -community user movie set and ensuring that the user and the movie do not repeatedly appear in the user movie set;
(g) and (4) updating the k and s values according to the index table entry newly found at present, repeating the steps (c) to (g) until the compression direction in the accessed compression table entry is null, and stopping iteration.
4. A server comprising a processor and a memory, the memory having stored therein at least one instruction, at least one program, set of codes, or set of instructions, which is loaded and executed by the processor to implement the method of any one of claims 1 to 3.
5. A computer readable storage medium having stored therein at least one instruction, at least one program, set of codes, or set of instructions, which is loaded and executed by a processor to implement the method of any one of claims 1 to 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111142284.4A CN113806612B (en) | 2021-09-28 | 2021-09-28 | Method for detecting key communities in user movie network based on index |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111142284.4A CN113806612B (en) | 2021-09-28 | 2021-09-28 | Method for detecting key communities in user movie network based on index |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113806612A true CN113806612A (en) | 2021-12-17 |
| CN113806612B CN113806612B (en) | 2024-01-02 |
Family
ID=78938702
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111142284.4A Active CN113806612B (en) | 2021-09-28 | 2021-09-28 | Method for detecting key communities in user movie network based on index |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113806612B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100138443A1 (en) * | 2008-11-17 | 2010-06-03 | Ramakrishnan Kadangode K | User-Powered Recommendation System |
| CN103793476A (en) * | 2014-01-08 | 2014-05-14 | 西安电子科技大学 | Network community based collaborative filtering recommendation method |
| US20160117414A1 (en) * | 2014-10-23 | 2016-04-28 | Sudhir Verma | In-Memory Database Search Optimization Using Graph Community Structure |
| CN111666468A (en) * | 2020-05-11 | 2020-09-15 | 浙江工商大学 | Method for searching personalized influence community in social network based on cluster attributes |
-
2021
- 2021-09-28 CN CN202111142284.4A patent/CN113806612B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100138443A1 (en) * | 2008-11-17 | 2010-06-03 | Ramakrishnan Kadangode K | User-Powered Recommendation System |
| CN103793476A (en) * | 2014-01-08 | 2014-05-14 | 西安电子科技大学 | Network community based collaborative filtering recommendation method |
| US20160117414A1 (en) * | 2014-10-23 | 2016-04-28 | Sudhir Verma | In-Memory Database Search Optimization Using Graph Community Structure |
| CN111666468A (en) * | 2020-05-11 | 2020-09-15 | 浙江工商大学 | Method for searching personalized influence community in social network based on cluster attributes |
Non-Patent Citations (1)
| Title |
|---|
| 周小平;梁循;张海燕;: "基于R-C模型的微博用户社区发现", 软件学报, no. 12 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113806612B (en) | 2024-01-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108920720B (en) | Large-scale image retrieval method based on depth hash and GPU acceleration | |
| CN111797321B (en) | Personalized knowledge recommendation method and system for different scenes | |
| CN107833117B (en) | Bayesian personalized sorting recommendation method considering tag information | |
| CN112182245B (en) | Knowledge graph embedded model training method and system and electronic equipment | |
| CN103678672A (en) | Method for recommending information | |
| CN111563192B (en) | Entity alignment method, device, electronic equipment and storage medium | |
| CN105335368B (en) | A kind of product clustering method and device | |
| Altingovde et al. | Cluster searching strategies for collaborative recommendation systems | |
| CN113918833B (en) | Product recommendation method realized through graph convolution collaborative filtering of social network relationship | |
| WO2022142001A1 (en) | Target object evaluation method based on multi-score card fusion, and related device therefor | |
| CN113918832B (en) | Graph convolution collaborative filtering recommendation system based on social relationship | |
| CN111666468A (en) | Method for searching personalized influence community in social network based on cluster attributes | |
| CN114154565A (en) | Click rate prediction method and device based on multi-level feature interaction | |
| CN113761359A (en) | Data packet recommendation method and device, electronic equipment and storage medium | |
| CN113918834A (en) | Graph convolution collaborative filtering recommendation method fusing social relations | |
| CN115438169A (en) | Text and video mutual inspection method, device, equipment and storage medium | |
| CN113342994B (en) | Recommendation system based on non-sampling cooperative knowledge graph network | |
| Yasir et al. | D-GENE: deferring the GENEration of power sets for discovering frequent itemsets in sparse big data | |
| CN113806612A (en) | Method for detecting key community in user movie network based on index | |
| CN113744023B (en) | Double-channel collaborative filtering recommendation method based on graph convolution network | |
| CN114925268A (en) | Recommendation method and system based on graph neural network, electronic device and computer readable medium | |
| Zarzour et al. | An Efficient Recommender System Based on Collaborative Filtering Recommendation and Cluster Ensemble | |
| CN112817963B (en) | Community kernel decomposition method and system on multidimensional network | |
| CN116402589B (en) | Commodity recommendation method and system based on knowledge graph and rotary coding | |
| Chiu et al. | Efficiently processing skyline query on multi-instance data |
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 |