CN104794636B - The type for showing scoring based on user recommends method - Google Patents
The type for showing scoring based on user recommends method Download PDFInfo
- Publication number
- CN104794636B CN104794636B CN201510186309.9A CN201510186309A CN104794636B CN 104794636 B CN104794636 B CN 104794636B CN 201510186309 A CN201510186309 A CN 201510186309A CN 104794636 B CN104794636 B CN 104794636B
- Authority
- CN
- China
- Prior art keywords
- mrow
- mobile phone
- user
- msub
- type
- 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
Landscapes
- Mobile Radio Communication Systems (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The type for showing scoring based on user recommends method, comprises the following steps：1) mobile phone type Similarity Measure：A selects some mobile phone types of current main flow according to timing node, and the parameter information (such as system, internal memory etc.) of each mobile phone type is gathered from internet；B is pre-processed the data obtained in previous step, is deposited into after formatting in database, is produced mobile phone model information table mobile；C calculates the similarity between each mobile phone type, obtains mobile phone type similarity table mobile_sim according to mobile phone model information obtained in the previous step；D terminates；2) targeted customer's neighborhood solves the stage, realizes a kind of efficient mobile phone type and recommends method, to be preferably that mobile phone businessman carries out type marketing.
Description
Technical field
The present invention relates to a kind of mobile phone type to recommend method, i.e., is scored number using interconnecting display of the user on the network to mobile phone
According to the technology to user's progress mobile phone recommendation.
Background technology
Can be with intellectual analysis user's history behavioral data using computer technology, the concurrent current potential interest in family.Intelligence
The development of mobile phone and mobile Internet so that the frequency of the hand-off machine of user also more and more higher.Current either handset manufacturers, hand
Machine retailer or mobile operator carry out the recommendation of mobile phone type to user and compare concern.In conventional recommended technology, collaboration
Filtered recommendation is one of famous technology.But traditional collaborative filtering recommending technology based on user is calculating the similar of user
When spending, identical weights are all to confer to for every money mobile phone type, have ignored user's high scoring common to mobile phone and common low
The influence of difference and mobile phone monoid to recommendation results between scoring.So it is easy for causing to recommend efficiency low, is unfavorable for
Businessman carries out mobile phone marketing.The present invention attempts to utilize traditional collaborative filtering recommending thought based on user, and is carried for above-mentioned
To the problem of propose solution, design it is a kind of based on interconnection user on the network show score data mobile phone type recommendation side
Method.
The content of the invention
The present invention seeks to technical problem to be solved is to show the number to score to mobile phone using user on the network is interconnected
According to realizing a kind of efficient mobile phone type and recommend method, to be preferably that mobile phone businessman carries out type marketing.
To solve the above problems, the technical scheme is that, show that the type of scoring recommends method based on user, including
Following steps：
1) mobile phone type Similarity Measure：
A selects some mobile phone types of current main flow according to timing node, and each mobile phone type is gathered from internet
Parameter information (such as system, internal memory etc.)；
B is pre-processed the data obtained in previous step, is deposited into after formatting in database, is produced mobile phone machine
Type information table mobile；
C calculates the similarity between each mobile phone type, obtains mobile phone machine according to mobile phone model information obtained in the previous step
Type similarity table mobile_sim；
D terminates；
2) targeted customer's neighborhood solves the stage：
A gathers scoring of the user to mobile phone type from internet, obtains user-mobile phone evaluation information table user_
mobile；
B calculates the similarity between user and user using user's similarity formula, obtains user's similarity table user_
sim；
Neighborhood neighbor of the K similarity highest user as targeted customer a before c chooses；
D terminates；
3) mobile phone type recommendation list solves the stage：
A scans form user_mobile, therefrom extracts the unvalued type list T of user uu；
B is directed to list TuIn every Mobile phone type i, calculate user u to i interest-degree, obtain interest-degree list Pu；
C is by PuIt is ranked up, selects the maximum top n mobile phone type of interest-degree as the recommendation list to user u；
D terminates；
Described data prediction and formatting are that every money mobile phone type parameter information is expressed as into one in step 1)-b
Record, each record have several fields；
Mobile phone is mainly expressed as vector space model by described mobile phone type Similarity Measure in step 1)-c, then
Calculated using COS distance formula；
Step 2)-b detailed process is as follows：
1) scan round user-mobile phone evaluation information table, to wherein any two records：ua=＜ ra1,ra2,...rai＞
And ub=＜ rb1,rb2,...rbi＞, calculate the mobile phone type list T that user a and user b was evaluated jointlyab；
Wherein ua,ubRepresent user a and user b, rai,rbiRepresent to represent user a and user b to mobile phone type i's respectively
Score value；
2) sequential scan type list TabIn each type, solve its due weight w (i)=w1(i)×w2
(i), wherein
rai,rbiScore values of the user a and user b to mobile phone type i is represented respectively.The formula table
Bright, the mobile phone type weights of common high scoring are more than the weights of common lower assessment departure machine type between user；
hijRepresent mobile phone t in common scoring listiWith TabIn other mobile phones tjSimilarity, p
Represent TabThe number of middle type.The formula shows, the mobile phone in common list of scoring belong to same monoid (i.e. similarity compared with
It is high) mobile phone is more, and the weights of the mobile phone are bigger；
3) user a and user b similarity is calculated by calculating formula of similarity, the formula of calculating is：
Wherein
Described preceding K similarity highest user in step 2)-c, it is the user u come will to be solved in step 2)-b
A sequence is carried out with the similarity of other users, neighborhood of the K user as a before then selecting, K value can basis
It is actual to promote result situation to carry out constantly adjustment to obtain a best empirical value K；
4) terminate；
User u is to mobile phone type i interest-degree calculation formula in step 3)-b：
Wherein,Represent user u and user b to having commented the scoring average of mobile phone type respectively；
Described selection top n mobile phone type in step 3)-c, N value can promote result situation to carry out not according to actual
Disconnected adjustment is to obtain a best empirical value N；
Beneficial effect of the present invention：The present invention shows that the mobile phone type of score data recommends method solution based on interconnection user on the network
The deficiency determined in traditional filtering recommendation algorithms based on user collaborative, not only consider the common high scoring of user and common low
Difference between scoring, and have also contemplated that influence of the mobile phone monoid to recommendation results.The height of recommendation results can so be improved
Effect property, mobile phone businessman is helped preferably to be marketed.
Brief description of the drawings
Fig. 1 is that mobile phone type recommends operational flowchart.
Fig. 2 is the flow chart of the mobile phone type recommendation method that score data is shown based on interconnection user on the network of the present invention.
Fig. 3 is user's neighborhood solution procedure flow chart.
Fig. 4 is that mobile phone type recommendation list produces phase flow figure.
Embodiment
In order to know more about the technology contents of the present invention, especially exemplified by specific embodiment and institute's accompanying drawings are coordinated to be described as follows.
As shown in figure 1, mobile phone type is recommended mainly to show score data and mobile phone type by obtaining interconnection user on the network
Several stages such as parameter information, data prediction and formatting, user's neighborhood solve, mobile phone type recommendation list produces.
The process is also based on the main thought in the collaborative filtering recommending method of user.
User's neighborhood solves and mobile phone recommendation list is main two stages of the process.The present invention thinking be exactly
The two steps are carried out with further processing, i.e., during the similarity of user is solved, it is contemplated that per money mobile phone type
Weights, rather than make no exception as traditional method and treat.The accuracy of recommendation results can so be improved.
The present invention's shows that the mobile phone type of score data recommends flow chart such as Fig. 2 institutes of method based on interconnection user on the network
Show.
Step 1 recommends the initial state of method for the mobile phone type of the present invention；
Step 2 is that (such as electric business website) obtains mobile phone type parameter information, user-mobile phone evaluation letter on the internet
Breath；
Step 3 is to calculate the similarity between different mobile phone types, typically uses COS distance formula；
Step 4 is to calculate the neighborhood of targeted customer；
Step 5 is to produce the mobile phone type recommendation list stage to targeted customer；
Step 6 is Fig. 2 end；
Fig. 3 is the detailed description to step 4 in Fig. 2.
Step 7 is initial state；
Step 8 input needs the user u to its recommending mobile phone type；
Step 9 definition and the similarity array s between initialising subscriber u and other users v；
Step 10 is each record in cyclic access user-mobile phone evaluation information table, then operating procedure 11-16,
Untill accessing completion；
Step 11 is to calculate the mobile phone type list T that targeted customer scored jointly with other usersuv；
Step 12 sequential access TuvIn every Mobile phone type i；
Step 13 calculates the due weight ws (i) of i；
Step 14 calculates weighted scoring average m (u；And m (v w)；w)；
Step 15 calculates the similarity sim (uv) between user u and user v；
The user v and its similarity sim (uv) are added in s by step 16；
Step 17 will be ranked up from big to small using a kind of sort method in s according to sim (uv)；
Neighbours' number K that step 18 input needs；
Neighborhood of the K user as u before step 19 selects in s；
Step 20 is done state；
Fig. 4 is the detailed description to step 5 in Fig. 2.
Step 21 is initial state；
Step 22 input needs the user u to its recommending mobile phone type；
Step 23 definition and the initialising subscriber u and interest-degree array P per money mobile phone i；
Step 24 sequential access user_mobile tables, solve the unvalued mobile phone type list M of user u；
Every Mobile phone type in step 25 cyclic access M, and carry out step 26-27 operation；
Step 26 calculates interest-degree ps (u, i) of the user u to mobile phone type i；
Mobile phone type i and corresponding p (u, i) are added in array P by step 27；
P is ranked up by step 28 from big to small according to p (u, i)；
Step 29 input needs the type number N recommended；
Step 30 selects top n mobile phone type as consequently recommended list in P；
Step 31 is done state.
In summary, present invention utilizes the collaborative filtering recommending method thought based on user, and opponent between user is considered
The common high scoring of machine and common lower assessment subregion are not and influence of the monoid to Similarity Measure between mobile phone, traditional so as to solve
In method calculate similarity when be all the defects for the treatment of of making no exception to different mobile phones.This quadrat method can effectively be improved and pushed away
The degree of accuracy of result is recommended, providing marketing for mobile phone businessman supports.
Claims (2)
1. the type for showing scoring based on user recommends method, it is characterized in that comprising the following steps：
1) mobile phone type Similarity Measure：
A selects some mobile phone types of current main flow according to timing node, and the bag of each mobile phone type is gathered from internet
Include system, the parameter information of internal memory；
B is pre-processed the data obtained in previous step, is deposited into after formatting in database, produces mobile phone type letter
Cease table mobile；
C calculates the similarity between each mobile phone type, obtains mobile phone type phase according to mobile phone model information obtained in the previous step
Like degree table mobile_sim；
D terminates；
2) targeted customer's neighborhood solves the stage：
A gathers scoring of the user to mobile phone type from internet, obtains user-mobile phone evaluation information table user_mobile；
B calculates the similarity between user and user using user's similarity formula, obtains user's similarity table user_sim；
Neighborhood neighbor of the K similarity highest user as targeted customer a before c chooses；
D terminates；
3) mobile phone type recommendation list solves the stage：
A scans form user_mobile, therefrom extracts the unvalued type list T of user uu；
B is directed to list TuIn every Mobile phone type i, calculate user u to i interest-degree, obtain interest-degree list Pu；
C is by PuIt is ranked up, selects the maximum top n mobile phone type of interest-degree as the recommendation list to user u；
D terminates；
Described data prediction and formatting are that every money mobile phone type parameter information is expressed as into a record in step 1)-b,
Each record has several fields；
Described mobile phone type Similarity Measure is that mobile phone is expressed as into vector space model in step 1)-c, then utilizes cosine
Range formula is calculated；
Step 2)-b detailed process is as follows：
1) scan round user-mobile phone evaluation information table, to wherein any two records：ua=＜ ra1,ra2,...rai＞ and ub
=＜ rb1,rb2,...rbi＞, calculate the mobile phone type list T that user a and user b was evaluated jointlyab；
Wherein ua,ubRepresent user a and user b, rai,rbiScore values of the user a and user b to mobile phone type i is represented respectively；
2) sequential scan type list TabIn each type, solve its due weight w (i)=w1(i)×w2(i), wherein
rai,rbiScore values of the user a and user b to mobile phone type i is represented respectively；The formula shows, uses
The mobile phone type weights of common high scoring are more than the weights of common lower assessment departure machine type between family；
hijRepresent mobile phone t in common scoring listiWith TabIn other mobile phones tjSimilarity, p represent
TabThe number of middle type；The formula shows that the mobile phone in common list of scoring belongs to the i.e. similarity of same monoid compared with high mobile phone
More, the weights of the mobile phone are bigger；
3) user a and user b similarity is calculated by calculating formula of similarity, the formula of calculating is：
<mrow>
<mi>S</mi>
<mi>i</mi>
<mi>m</mi>
<mrow>
<mo>(</mo>
<mi>a</mi>
<mo>,</mo>
<mi>b</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mfrac>
<mrow>
<munder>
<mo>&Sigma;</mo>
<mrow>
<msub>
<mi>t</mi>
<mi>i</mi>
</msub>
<mo>&Element;</mo>
<msub>
<mi>T</mi>
<mrow>
<mi>a</mi>
<mi>b</mi>
</mrow>
</msub>
</mrow>
</munder>
<mi>w</mi>
<mrow>
<mo>(</mo>
<mi>i</mi>
<mo>)</mo>
</mrow>
<mrow>
<mo>(</mo>
<msub>
<mi>r</mi>
<mrow>
<mi>a</mi>
<mi>i</mi>
</mrow>
</msub>
<mo>-</mo>
<mi>m</mi>
<mo>(</mo>
<mrow>
<mi>a</mi>
<mo>;</mo>
<mi>w</mi>
</mrow>
<mo>)</mo>
<mo>)</mo>
</mrow>
<mrow>
<mo>(</mo>
<msub>
<mi>r</mi>
<mrow>
<mi>b</mi>
<mi>i</mi>
</mrow>
</msub>
<mo>-</mo>
<mi>m</mi>
<mo>(</mo>
<mrow>
<mi>b</mi>
<mo>;</mo>
<mi>w</mi>
</mrow>
<mo>)</mo>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<msqrt>
<mrow>
<munder>
<mo>&Sigma;</mo>
<mrow>
<msub>
<mi>t</mi>
<mi>i</mi>
</msub>
<mo>&Element;</mo>
<msub>
<mi>T</mi>
<mrow>
<mi>a</mi>
<mi>b</mi>
</mrow>
</msub>
</mrow>
</munder>
<mi>w</mi>
<mrow>
<mo>(</mo>
<mi>i</mi>
<mo>)</mo>
</mrow>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mi>r</mi>
<mrow>
<mi>a</mi>
<mi>i</mi>
</mrow>
</msub>
<mo>-</mo>
<mi>m</mi>
<mo>(</mo>
<mrow>
<mi>a</mi>
<mo>;</mo>
<mi>w</mi>
</mrow>
<mo>)</mo>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
</mrow>
</msqrt>
<msqrt>
<mrow>
<munder>
<mo>&Sigma;</mo>
<mrow>
<msub>
<mi>t</mi>
<mi>i</mi>
</msub>
<mo>&Element;</mo>
<msub>
<mi>T</mi>
<mrow>
<mi>a</mi>
<mi>b</mi>
</mrow>
</msub>
</mrow>
</munder>
<mi>w</mi>
<mrow>
<mo>(</mo>
<mi>i</mi>
<mo>)</mo>
</mrow>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mi>r</mi>
<mrow>
<mi>b</mi>
<mi>i</mi>
</mrow>
</msub>
<mo>-</mo>
<mi>m</mi>
<mo>(</mo>
<mrow>
<mi>b</mi>
<mo>;</mo>
<mi>w</mi>
</mrow>
<mo>)</mo>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
</mrow>
</msqrt>
</mrow>
</mfrac>
<mo>,</mo>
</mrow>
Wherein
Described preceding K similarity highest user in step 2)-c, it is next user u and its will to be solved in step 2)-b
The similarity of his user carries out a sequence, and neighborhood of the K user as a before then selecting, K value can be according to reality
Result situation is promoted to carry out constantly adjustment to obtain a best empirical value K；
4) terminate.
2. according to claim 1 show that the type of scoring recommends method based on user, it is characterized in that
User u is to mobile phone type i interest-degree calculation formula in step 3)-b：
<mrow>
<mi>p</mi>
<mrow>
<mo>(</mo>
<mi>u</mi>
<mo>,</mo>
<mi>i</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mover>
<msub>
<mi>r</mi>
<mi>u</mi>
</msub>
<mo>&OverBar;</mo>
</mover>
<mo>+</mo>
<mfrac>
<mrow>
<munder>
<mo>&Sigma;</mo>
<mrow>
<mi>b</mi>
<mo>&Element;</mo>
<mi>n</mi>
<mi>e</mi>
<mi>i</mi>
<mi>g</mi>
<mi>h</mi>
<mi>b</mi>
<mi>o</mi>
<mi>r</mi>
</mrow>
</munder>
<mi>S</mi>
<mi>i</mi>
<mi>m</mi>
<mrow>
<mo>(</mo>
<mi>u</mi>
<mo>,</mo>
<mi>b</mi>
<mo>)</mo>
</mrow>
<mo>*</mo>
<mrow>
<mo>(</mo>
<msub>
<mi>r</mi>
<mrow>
<mi>b</mi>
<mo>,</mo>
<mi>i</mi>
</mrow>
</msub>
<mo>-</mo>
<mover>
<msub>
<mi>r</mi>
<mi>b</mi>
</msub>
<mo>&OverBar;</mo>
</mover>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<munder>
<mo>&Sigma;</mo>
<mrow>
<mi>b</mi>
<mo>&Element;</mo>
<mi>n</mi>
<mi>e</mi>
<mi>i</mi>
<mi>g</mi>
<mi>h</mi>
<mi>b</mi>
<mi>o</mi>
<mi>r</mi>
</mrow>
</munder>
<mo>|</mo>
<mi>S</mi>
<mi>i</mi>
<mi>m</mi>
<mrow>
<mo>(</mo>
<mi>u</mi>
<mo>,</mo>
<mi>b</mi>
<mo>)</mo>
</mrow>
<mo>|</mo>
</mrow>
</mfrac>
</mrow>
Wherein,Represent user u and user b to having commented the scoring average of mobile phone type respectively；
Described selection top n mobile phone type in step 3)-c, N value can promote result situation to carry out constantly according to actual
Adjust to obtain a best empirical value N.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510186309.9A CN104794636B (en) | 2015-04-17 | 2015-04-17 | The type for showing scoring based on user recommends method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510186309.9A CN104794636B (en) | 2015-04-17 | 2015-04-17 | The type for showing scoring based on user recommends method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104794636A CN104794636A (en) | 2015-07-22 |
| CN104794636B true CN104794636B (en) | 2018-04-06 |
Family
ID=53559420
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510186309.9A Active CN104794636B (en) | 2015-04-17 | 2015-04-17 | The type for showing scoring based on user recommends method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104794636B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106503022B (en) * | 2015-09-08 | 2020-12-01 | 北京邮电大学 | Method and device for pushing recommendation information |
| CN105488684A (en) * | 2015-11-16 | 2016-04-13 | 孙宝文 | Method and apparatus for determining recommendation relationship in trading system |
| CN107016460A (en) * | 2017-03-27 | 2017-08-04 | 中国联合网络通信集团有限公司广西壮族自治区分公司 | User changes planes Forecasting Methodology and device |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101556615A (en) * | 2008-04-08 | 2009-10-14 | 索尼株式会社 | Information processing apparatus and method, recommendation apparatus and method, and information processing system |
| CN103064924A (en) * | 2012-12-17 | 2013-04-24 | 浙江鸿程计算机系统有限公司 | Travel destination situation recommendation method based on geotagged photo excavation |
| CN103514304A (en) * | 2013-10-29 | 2014-01-15 | 海南大学 | Project recommendation method and device |
| CN104008515A (en) * | 2014-06-04 | 2014-08-27 | 江苏金智教育信息技术有限公司 | Intelligent course selection recommendation method |
| CN104394162A (en) * | 2014-12-03 | 2015-03-04 | 南京智数科技有限公司 | Service quality credibility considered network service selection method |
-
2015
- 2015-04-17 CN CN201510186309.9A patent/CN104794636B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101556615A (en) * | 2008-04-08 | 2009-10-14 | 索尼株式会社 | Information processing apparatus and method, recommendation apparatus and method, and information processing system |
| CN103064924A (en) * | 2012-12-17 | 2013-04-24 | 浙江鸿程计算机系统有限公司 | Travel destination situation recommendation method based on geotagged photo excavation |
| CN103514304A (en) * | 2013-10-29 | 2014-01-15 | 海南大学 | Project recommendation method and device |
| CN104008515A (en) * | 2014-06-04 | 2014-08-27 | 江苏金智教育信息技术有限公司 | Intelligent course selection recommendation method |
| CN104394162A (en) * | 2014-12-03 | 2015-03-04 | 南京智数科技有限公司 | Service quality credibility considered network service selection method |
Non-Patent Citations (2)
| Title |
|---|
| 一个基于信任网络的推荐系统研究与应用;王宁;《中国优秀硕士学位论文全文数据库 信息科技辑》;20141215;20-25 * |
| 基于用户相似性的协同过滤推荐算法研究;程飞;《中国优秀硕士学位论文全文数据库 信息科技辑》;20121015;34-37 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104794636A (en) | 2015-07-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102841946B (en) | Commodity data retrieval ordering and Method of Commodity Recommendation and system | |
| CN103745100B (en) | A kind of method of the collaborative filtering recommending of the dominant explicit feedback of project-based mixing | |
| CN107330750B (en) | A kind of recommended products figure method and device, electronic equipment | |
| CN103886048B (en) | Cluster-based increment digital book recommendation method | |
| CN102063433A (en) | Method and device for recommending related items | |
| CN106447388A (en) | Method and system for recommending dishes | |
| CN103761237A (en) | Collaborative filtering recommending method based on characteristics and credibility of users | |
| CN103942712A (en) | Product similarity based e-commerce recommendation system and method thereof | |
| CN105354260B (en) | The Mobile solution of a kind of mosaic society's network and item characteristic recommends method | |
| CN103428165B (en) | Method and device for grouping social network nodes | |
| CN103412937A (en) | Searching and shopping method based on handheld terminal | |
| CN104794636B (en) | The type for showing scoring based on user recommends method | |
| CN103164424B (en) | Method and device for acquiring time-efficient words | |
| CN104182543B (en) | Recommend method with the mixing of popularity dimensionality reduction based on similar spreading through sex intercourse | |
| CN102982466B (en) | A kind of score in predicting method based on user's liveness | |
| CN107944035A (en) | A kind of image recommendation method for merging visual signature and user's scoring | |
| CN108804683A (en) | Associate(d) matrix decomposes and the film of collaborative filtering recommends method | |
| CN102779157B (en) | Method and device for searching images | |
| CN106021329A (en) | A user similarity-based sparse data collaborative filtering recommendation method | |
| CN103353920A (en) | Method and device for recommending games based on SNS | |
| CN109902235A (en) | User preference based on bat optimization clusters Collaborative Filtering Recommendation Algorithm | |
| CN105023178B (en) | A kind of electronic commerce recommending method based on ontology | |
| CN104715127B (en) | One kind complains hot spot region recognition methods and system | |
| CN103744904B (en) | A kind of method and device that information is provided | |
| CN106708883A (en) | Recommendation method and apparatus |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| EXSB | Decision made by sipo to initiate substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |