CN110569435B - Intelligent dual-ended recommendation engine system and method - Google Patents
Intelligent dual-ended recommendation engine system and method Download PDFInfo
- Publication number
- CN110569435B CN110569435B CN201910805151.7A CN201910805151A CN110569435B CN 110569435 B CN110569435 B CN 110569435B CN 201910805151 A CN201910805151 A CN 201910805151A CN 110569435 B CN110569435 B CN 110569435B
- Authority
- CN
- China
- Prior art keywords
- dimension
- portrait
- user
- product
- data
- 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
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
Abstract
The invention relates to the field of big data and accurate pushing, and discloses an intelligent double-end recommendation engine system, which comprises a data system, a decision system, a dimension system, a rule system and a mirroring system, wherein the mirroring system is used for carrying out back-stepping to form the dimension of an expected portrait at a product end; the system can realize double-end matching and accurate pushing by using the characteristics of thousands of users in software products at a large number of user ends and a large number of product ends in a double-end portrait technical mode, so that the user can find the most suitable software product among a plurality of similar products, and the safety of personal capital and property can be ensured at the same time; meanwhile, the end software product of the product end can also reduce the wind control cost of the product end and improve the accurate popularization of the brand of the product end.
Description
Technical Field
The invention relates to the field of big data and accurate pushing, in particular to an intelligent double-end recommendation engine system and method.
Background
The rapid development of the internet software industry promotes various industries to have large-batch similar software products (namely B-side). A user (namely C end) needs to find a most suitable B end software product among a plurality of similar products, and simultaneously, the safety of personal capital and property needs to be ensured; the B-end software product also hopes to reduce the own wind control cost, improve the accurate popularization of the brand of the own product, and further hopes to make the own product location and applicable group clear. Both sides desire a convenient and reliable finding channel which can identify and avoid risks in advance. How to help both ends solve such selection problems has been increasingly appreciated.
At present, engine products with a recommendation function in the market mostly perform a cartesian product on a C-end user and a B-end product, and the precision is almost 0 without good and good full-scale pushing, so that a large amount of resource waste is caused, the use experience of the C-end and the B-end is very poor, and particularly, the extremely high trial-and-error rate of the C-end user is caused. More importantly, the Cartesian product type push increases the asset risk of the C-end user. Taking the internet financial industry as an example, once a C-end user mistakenly enters a trap with high default and high profit rate under malicious induction, a very serious social problem can be caused.
Even if a small amount of engine products capable of classifying and positioning the C-end users can only achieve single-end matching aiming at the B-end products, namely, the users are pushed to the B-end according to the portrait of the C-end users, and the requirements of the C-end users are not met: the most safe, best quality and most suitable product for the user is selected from a plurality of similar products.
Disclosure of Invention
The invention solves the technical problem of providing an intelligent double-end recommendation engine system, which realizes double-end matching and accurate pushing by using the characteristics of thousands of users and faces in a double-end portrait technical mode from a large number of user end users and a large number of product end software products.
The technical scheme adopted by the invention for solving the technical problems is as follows: an intelligent dual ended recommendation engine system comprising:
the data system comprises a user side data module for storing user side data and a product side data module for storing product side data;
the decision system is used for being responsible for dimension switching, allocation work and dimension judgment result analysis, and determining the flow direction according to the judgment result of each node;
the dimension system is used for carrying out analysis calculation according to the data stored in the data system and a set dimension standard to obtain a final dimension value and feeding the final dimension value back to the decision-making system;
the rule system is used for rule configuration, dimension judgment and dimension grading;
a mirroring system for retrograding to form product end desired portrait dimensions;
the portrait generation system determines all the dimensional characteristics of the two ends through dimensional values and dimensional judgment results calculated and summarized by a decision system, a rule system and a mirroring system, then sorts and merges large data, and finally generates self portrait and expected portrait of a user end and a product end through symmetric encryption;
a representation storage system for storing self representations of the user side and the product side generated in the representation generation system and a desired representation;
a portrait matching recommendation system for recommending the closest self portrait and the desired portrait to each other;
the data system, the decision system, the dimension system, the rule system, the mirroring system, the portrait generation system, the portrait storage system and the portrait matching recommendation system are connected and interacted with one another.
Further, the method comprises the following steps: the data sources stored by the user side data module and the product side data module comprise filled basic information, equipment information data acquired under authorization and operator and other authoritative third party credit investigation data inquired after authorization.
Further, the method comprises the following steps: the dimensions include a base dimension, a label dimension, and an integrated dimension; the sketch matching recommendation system comprises fuzzy matching, touch matching and accurate matching;
the fuzzy matching satisfies that the basic dimension matching degree of the user side and the product side reaches A%, and the dimension matching rate of the user side and the product side reaches B%;
the touch matching meets the fuzzy matching, and meanwhile, the comprehensive dimension matching rate of the user side and the product side reaches more than C%;
the accurate matching meets the fuzzy matching, and meanwhile, the comprehensive dimension matching rate of the user side and the product side reaches more than D%, wherein D is larger than C.
Further, the method comprises the following steps: the product end expected portrait dimension reverse-deducing step comprises the following steps:
the method comprises the following steps: acquiring the existing full dimensionality and the user data using the product end;
step two: acquiring a dimension;
step three: acquiring user data of the product end;
step four: judging whether the user has the dimension value, if not, marking the user as empty, if so, marking the user as yes or no, and marking the user as score value by the dimension judgment type mark;
step five: judging whether a next user exists, if so, returning to the step three, and if not, entering the step six;
step six: judging whether a next dimension exists, if so, returning to the step two, and if not, entering the step seven;
step seven: acquiring all marked dimensions;
step eight: taking one dimension in the step seven;
step nine: if the dimension is the dimension judgment type, inquiring the mark with the most dimension proportion, if the mark proportion is more than 80%, keeping the dimension and the mark value, otherwise, abandoning the dimension, and if the dimension is the dimension evaluation type, marking the scoring interval in which 80% of the top dimension is positioned;
step ten: judging whether a next marked dimension exists, if so, returning to the step eight, and if not, entering the step eleven;
step eleven: and sorting the dimensions of the dimension judgment class and the dimension evaluation class and the marking values thereof, generating a desired portrait through a portrait factory, and storing the expected portrait in a portrait library.
Further, the method comprises the following steps: the user terminal self portrait generation step:
the method comprises the following steps: obtaining users waiting to make pictures in batch;
step two: acquiring a user;
step three: acquiring all process nodes;
step four: taking a node;
step five: inquiring all dimensions under the node, reading user data in a user side data module, calculating dimension values according to set logics, recording all dimension values under the node, then carrying out dimension judgment and dimension grading according to rules preset in a rule system, and recording rule results of all dimensions under the node;
step six: judging whether a next node exists, if so, returning to the step four, and if not, entering the step seven;
step seven: sorting all dimension values, dimension judgment and dimension grading of the user, analyzing and encrypting the dimension values and the dimension grading, and finally generating a portrait and storing the portrait in a portrait library;
step eight: and C, judging whether a next user exists or not, if so, returning to the step two, and if not, ending the whole process.
The invention also discloses an intelligent double-end recommendation method, which comprises the following steps:
the method comprises the following steps: acquiring various original data of a user side and a product side;
step two: generating self portrait and expected portrait of a user terminal through a decision system, a dimension system, a rule system, a mirror system and a portrait generation system, and storing the self portrait and the expected portrait in a portrait storage system;
step three: a portrait matching recommendation system recommends the closest self portrait and the desired portrait stored in the portrait storage system to each other.
The invention has the beneficial effects that: the system can realize double-end matching and accurate pushing by using the characteristics of thousands of faces of thousands of users in a double-end portrait technical mode from a large number of user end users and a large number of product end software products, so that the users can find the most suitable software products among a plurality of similar products, and the personal capital and property safety can be ensured at the same time; meanwhile, the end software product of the product end can also reduce the wind control cost of the product end and improve the accurate popularization of the brand of the product end.
Drawings
FIG. 1 is a flow chart of an accurate recommendation platform of the present invention;
FIG. 2 is an exemplary diagram of a rule writing of the rule system of the present invention;
FIG. 3 is a flow diagram of user representation generation in accordance with the present invention;
FIG. 4 is a flow chart of desired representation generation according to the present invention.
Detailed Description
The invention is further described with reference to the following figures and detailed description.
The intelligent dual-ended recommendation engine system shown in fig. 1 comprises the following modules:
the data system comprises a user side data module for storing user side data and a product side data module for storing product side data;
the decision system is used for being responsible for dimension switching, allocation work and dimension judgment result analysis, and determining the flow direction according to the judgment result of each node; multiple decision processes can be set simultaneously to satisfy the calculation of multiple types of images,
the dimensionality system is used for carrying out analysis calculation according to the data stored in the data system and a set dimensionality standard to obtain a final dimensionality value and feeding the final dimensionality value back to the decision-making system;
the rule system is used for rule configuration, dimension judgment and dimension scoring;
a mirroring system for retrograding a product end desired portrait dimension;
the portrait generation system determines all the dimensional characteristics of the two ends through dimensional values and dimensional judgment results calculated and summarized by a decision system, a rule system and a mirroring system, then sorts and merges large data, and finally generates self portrait and expected portrait of a user end and a product end through symmetric encryption;
a representation storage system for storing self representations of the user side and the product side generated in the representation generation system and a desired representation;
a portrait matching recommendation system for recommending the closest self portrait and the desired portrait to each other;
the data system, the decision system, the dimension system, the rule system, the mirroring system, the portrait generation system, the portrait storage system and the portrait matching recommendation system are connected and interacted with one another.
The specific recommendation steps are as follows:
the method comprises the following steps: acquiring various original data of a user side and a product side;
step two: generating self portrait and expected portrait of a user terminal through a decision system, a dimension system, a rule system, a mirroring system and a portrait generating system, and storing the self portrait and the expected portrait in a portrait storage system;
step three: a portrait matching recommendation system recommends the closest self portrait and the desired portrait stored in the portrait storage system to each other.
Specifically, the data sources stored in the data system include filled basic information, device information data acquired under authorization, and operator and other authoritative third party credit investigation data inquired after authorization, and the basic information may be personal basic information of a user, the user's own requirements on functions and services of a product, a full-scale function point of the product, product user information, and the like, and further include a system internal blacklist, a white list, and the like of the own party.
On the basis, as shown in fig. 1, the dimensions include a base dimension, a label dimension and an integrated dimension; the sketch matching recommendation system comprises fuzzy matching, touch matching and accurate matching; the basic dimension comprises user basic information and basic data dimensions of both ends, such as user age, gender, school calendar and the like; the label dimension is a series of social characteristics of both ends, such as individual preference labels, position labels, industry labels of products and other label dimensions of different angles; a synthetic dimension, wherein all dimensions except the basic dimension and the label dimension belong to the synthetic dimension; the fuzzy matching meets the condition that the dimensionality matching degree of the user side and the product side reaches A%, and the dimensionality matching rate of the user side and the product side reaches B%; the touch matching meets the fuzzy matching, and meanwhile, the comprehensive dimension matching rate of the user side and the product side reaches more than C%; the accurate matching satisfies the fuzzy matching, and meanwhile, the comprehensive dimension matching rate of the user side and the product side reaches more than D%, wherein D is greater than C, and the thresholds of the three gradients can be configured and adjusted according to actual conditions, in the embodiment of the application, a =100, b =80, C =50, D =80.
On the basis, one rule in the rule system may include one or more dimensions, the determination condition includes but is not limited to size, non-null, whether to include a character string, whether to include in a set, and the like, and an intermediate result of each rule may be used as an entry of another rule. A rule set, a decision tree and a scoring card can be formed among the multiple rules. And combining a plurality of modes to finally realize the formulation of complex rule logic.
Meanwhile, the rule system realizes the safe editing of the rules. After the rule editor completes the addition, deletion and modification of the rule, the rule editor needs to be audited by auditors and then issued by the issuing personnel. Each rule change is processed by at least three roles together to complete formal release work, so that risks brought by wrong operation are effectively avoided, and the safety of the rule is ensured, as shown in fig. 2, the product is suitable for users between ages 40 and 50, the embodiment scores the gender and age of the user in the basic dimension, and the specific rule is as follows: firstly, judging whether the user has gender data, and if not, giving a score of 0; if the data exists, judging whether the user has age data, and if not, giving a score of 0; and scoring the age data with age data, wherein when the age is less than 18, a score is given for 0, when the age is greater than or equal to 18 and less than 30, a score is given for 15, when the age is greater than or equal to 30 and less than 40, a score is given for 25, when the age is greater than or equal to 40 and less than 50, a score is given for 30, when the age is greater than or equal to 50 and less than 65, a score is given for 20, when the age is greater than 65, a score is given for 10, and other dimension scoring rules can be written according to the actual condition.
In addition, as shown in fig. 3, the user-side self-portrait generating step:
the method comprises the following steps: obtaining users waiting to make pictures in batch;
step two: acquiring a user;
step three: acquiring all process nodes;
step four: taking a node;
step five: inquiring all dimensions under the node, reading user data in a user side data module, calculating dimension values according to set logics, recording all dimension values under the node, then carrying out dimension judgment and dimension grading according to rules preset in a rule system, and recording rule results of all dimensions under the node;
step six: judging whether a next node exists, if so, returning to the step four, and if not, entering the step seven;
step seven: sorting all dimension values, dimension judgment and dimension grading of the user, analyzing and encrypting the dimension values and the dimension grading, and finally generating a portrait and storing the portrait in a portrait library;
step eight: judging whether a next user exists, if so, returning to the step two, and if not, ending the whole process;
in the above steps, the whole process is composed of each node. Each node is for performing a certain periodic task, such as: one node is called as 'external blacklist verification', namely whether a certain user is in a blacklist of a third party is verified through third party data, a plurality of similar third party data mechanisms are provided, each blacklist of the third party is a dimension, if the certain user hits the blacklist of the A party, the dimension value of the A _ blacklist is 'yes', and if the certain user does not hit the blacklist of the B party, the dimension value of the B _ blacklist is 'no'.
On the basis, as shown in fig. 4, the product end expected image dimension backward-deducing step includes:
the method comprises the following steps: acquiring the existing full dimensionality and the user data using the product end;
step two: acquiring a dimension;
step three: acquiring user data of the product end;
step four: judging whether the user has the dimension value, if not, marking the user as empty, if so, marking the user as yes or no, and marking the user as score value by the dimension judgment type mark;
step five: judging whether a next user exists, if so, returning to the step three, and if not, entering the step six;
step six: judging whether a next dimension exists, if so, returning to the step two, and if not, entering the step seven;
step seven: acquiring all marked dimensions;
step eight: taking one dimension in the step seven;
step nine: if the dimension is the dimension judgment class, inquiring the mark with the largest dimension proportion, if the mark proportion is more than 80%, keeping the dimension and the mark value, otherwise, abandoning the dimension, and if the dimension is the dimension evaluation class, marking the scoring interval in which 80% of the first dimension is positioned;
step ten: judging whether a next marked dimension exists, if so, returning to the step eight, and if not, entering the step eleven;
step eleven: sorting the dimensions of the dimension judgment class and the dimension evaluation class and the marking values thereof, generating an expected portrait through a portrait factory, and storing the expected portrait in a portrait library;
the above-mentioned dimension judgment class returns a value of "yes" or "no", and the dimension evaluation class returns a score, for example: dimension determination class: is greater than 18 years old? The result is only "yes" or "no"; and dimension evaluation classification: the monthly income is less than 3600 and 60 minutes; 75 minutes between 3600 and 5000; 90 minutes between 5000 and 10000; 10000 or more and 100 points.
The system can realize double-end matching and accurate pushing by using the characteristics of thousands of faces of thousands of users in a double-end portrait technical mode from a large number of user end users and a large number of product end software products, so that the users can find the most suitable software products among a plurality of similar products, and simultaneously, the personal capital and property safety can be ensured; meanwhile, the end software product of the product end can also reduce the wind control cost of the product end and improve the accurate popularization of the brand of the product end.
The above-mentioned embodiments are intended to illustrate the objects, technical solutions and advantages of the present invention in further detail, and it should be understood that the above-mentioned embodiments are only exemplary embodiments of the present invention, and are not intended to limit the present invention, and any modifications, equivalents, improvements and the like made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (6)
1. An intelligent dual ended recommendation engine system, comprising:
the data system comprises a user side data module for storing user side data and a product side data module for storing product side data;
the decision system is used for being responsible for dimension switching, allocation work and dimension judgment result analysis, and determining the flow direction according to the judgment result of each node;
the dimensionality system is used for carrying out analysis calculation according to the data stored in the data system and a set dimensionality standard to obtain a final dimensionality value and feeding the final dimensionality value back to the decision-making system;
the rule system is used for rule configuration, dimension judgment and dimension grading;
a mirroring system for retrograding a product end desired portrait dimension;
the portrait generation system determines all the dimensional characteristics of the two ends through dimensional values and dimensional judgment results calculated and summarized by a decision system, a rule system and a mirroring system, then sorts and merges large data, and finally generates self portrait and expected portrait of a user end and a product end through symmetric encryption;
a representation storage system for storing self representations of the user side and the product side generated in the representation generation system and a desired representation;
a portrait matching recommendation system for recommending the closest self portrait and the desired portrait to each other;
the data system, the decision system, the dimension system, the rule system, the mirroring system, the portrait generation system, the portrait storage system and the portrait matching recommendation system are connected and interacted with one another.
2. The intelligent dual-end recommendation engine system as claimed in claim 1, wherein the data sources stored by the user-side data module and the product-side data module include filled basic information, device information data obtained under authorization, and credit data of operators and other authorized third parties queried after authorization.
3. The intelligent dual-ended recommendation engine system of claim 1, wherein the dimensions comprise a base dimension, a tag dimension, and a composite dimension; the figure matching recommendation system comprises fuzzy matching, touch matching and accurate matching;
the fuzzy matching meets the condition that the degree of matching between the basic dimensions of the user side and the product side reaches A%, and the degree of matching between the label dimensions of the user side and the product side reaches B%;
the touch matching meets the fuzzy matching, and meanwhile, the comprehensive dimension matching rate of the user side and the product side reaches more than C%;
the accurate matching meets the fuzzy matching, and meanwhile, the comprehensive dimension matching rate of the user side and the product side reaches more than D%, wherein D is larger than C.
4. The intelligent dual-ended recommendation engine system of claim 1, wherein said product-end desired portrait dimension back-deriving step comprises:
the method comprises the following steps: acquiring the existing full dimensionality and the user data using the product end;
step two: acquiring a dimension;
step three: acquiring user data of the product end;
step four: judging whether the user has the dimension value, if not, marking the user as empty, if so, marking the user as yes or no, and marking the user as score value by the dimension judgment type mark;
step five: judging whether a next user exists, if so, returning to the step three, and if not, entering the step six;
step six: judging whether a next dimension exists, if so, returning to the step two, and if not, entering the step seven;
step seven: acquiring all marked dimensions;
step eight: taking one dimension in the step seven;
step nine: if the dimension is the dimension judgment class, inquiring the mark with the largest dimension proportion, if the mark proportion is more than 80%, keeping the dimension and the mark value, otherwise, abandoning the dimension, and if the dimension is the dimension evaluation class, marking the scoring interval in which 80% of the first dimension is positioned;
step ten: judging whether a next marked dimension exists or not, if so, returning to the step eight, and if not, entering the step eleven;
step eleven: and sorting the dimensions of the dimension judgment class and the dimension evaluation class and the marking values thereof, generating a desired portrait through a portrait factory, and storing the expected portrait in a portrait library.
5. The intelligent dual-ended recommendation engine system of claim 1, wherein said user-side self-profile generating step:
the method comprises the following steps: obtaining users waiting to make pictures in batch;
step two: acquiring a user;
step three: acquiring all process nodes;
step four: taking a node;
step five: inquiring all dimensions under the node, reading user data in a user side data module, calculating dimension values according to set logics, recording all dimension values under the node, then carrying out dimension judgment and dimension grading according to rules preset in a rule system, and recording rule results of all dimensions under the node;
step six: judging whether a next node exists, if so, returning to the step four, and if not, entering the step seven;
step seven: sorting all dimension values, dimension judgment and dimension grading of the user, analyzing and encrypting the dimension values and the dimension grading, and finally generating a portrait and storing the portrait in a portrait library;
step eight: and C, judging whether a next user exists or not, if so, returning to the step two, and if not, ending the whole process.
6. An intelligent dual-ended recommendation method comprising the intelligent dual-ended recommendation engine system of any of claims 1-5, characterized by the steps of:
the method comprises the following steps: acquiring various original data of a user side and a product side;
step two: generating self portrait and expected portrait of a user terminal through a decision system, a dimension system, a rule system, a mirroring system and a portrait generating system, and storing the self portrait and the expected portrait in a portrait storage system;
step three: a portrait matching recommendation system recommends the closest self portrait and the desired portrait stored in the portrait storage system to each other.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910805151.7A CN110569435B (en) | 2019-08-29 | 2019-08-29 | Intelligent dual-ended recommendation engine system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910805151.7A CN110569435B (en) | 2019-08-29 | 2019-08-29 | Intelligent dual-ended recommendation engine system and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110569435A CN110569435A (en) | 2019-12-13 |
| CN110569435B true CN110569435B (en) | 2023-01-03 |
Family
ID=68776693
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910805151.7A Active CN110569435B (en) | 2019-08-29 | 2019-08-29 | Intelligent dual-ended recommendation engine system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110569435B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111062785A (en) * | 2019-12-18 | 2020-04-24 | 上海良鑫网络科技有限公司 | Method and system for intelligently selecting products to recommend to matched users |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010147269A1 (en) * | 2009-06-15 | 2010-12-23 | Cha Myoung Geun | Method and system for providing multifunctional search window service based on user-generated rules |
| CN104166713A (en) * | 2014-08-14 | 2014-11-26 | 百度在线网络技术（北京）有限公司 | Network service recommending method and device |
| CN107291841A (en) * | 2017-06-01 | 2017-10-24 | 广州衡昊数据科技有限公司 | A kind of method and system based on position and the social target of user's portrait intelligent Matching |
| CN108133407A (en) * | 2017-12-21 | 2018-06-08 | 湘南学院 | A kind of e-commerce recommended technology and system based on soft collection Decision Rule Analysis |
| CN109389520A (en) * | 2018-09-19 | 2019-02-26 | 国网山东省电力公司 | A kind of electric power system fault method for pushing and system |
| CN109783730A (en) * | 2019-01-03 | 2019-05-21 | 深圳壹账通智能科技有限公司 | Products Show method, apparatus, computer equipment and storage medium |
-
2019
- 2019-08-29 CN CN201910805151.7A patent/CN110569435B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010147269A1 (en) * | 2009-06-15 | 2010-12-23 | Cha Myoung Geun | Method and system for providing multifunctional search window service based on user-generated rules |
| CN104166713A (en) * | 2014-08-14 | 2014-11-26 | 百度在线网络技术（北京）有限公司 | Network service recommending method and device |
| CN107291841A (en) * | 2017-06-01 | 2017-10-24 | 广州衡昊数据科技有限公司 | A kind of method and system based on position and the social target of user's portrait intelligent Matching |
| CN108133407A (en) * | 2017-12-21 | 2018-06-08 | 湘南学院 | A kind of e-commerce recommended technology and system based on soft collection Decision Rule Analysis |
| CN109389520A (en) * | 2018-09-19 | 2019-02-26 | 国网山东省电力公司 | A kind of electric power system fault method for pushing and system |
| CN109783730A (en) * | 2019-01-03 | 2019-05-21 | 深圳壹账通智能科技有限公司 | Products Show method, apparatus, computer equipment and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 基于大数据技术的专家知识库设备画像推荐算法研究;王烨，郭玲利，宋文超，杨善友，程龙;《计算机测量与控制》;20181225;第26卷(第12期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110569435A (en) | 2019-12-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Winkler et al. | Automatic classification of requirements based on convolutional neural networks | |
| CN112613501A (en) | Information auditing classification model construction method and information auditing method | |
| CN107633380A (en) | The task measures and procedures for the examination and approval and system of a kind of anti-data-leakage system | |
| CN110826320A (en) | Sensitive data discovery method and system based on text recognition | |
| CN104573130A (en) | Entity resolution method based on group calculation and entity resolution device based on group calculation | |
| CN105787025A (en) | Network platform public account classifying method and device | |
| CN109800354B (en) | Resume modification intention identification method and system based on block chain storage | |
| CN111639291A (en) | Content distribution method, content distribution device, electronic equipment and storage medium | |
| CN107729549A (en) | A kind of robot client service method and system comprising elements recognition | |
| CN112966081A (en) | Method, device, equipment and storage medium for processing question and answer information | |
| CN111680506A (en) | External key mapping method and device of database table, electronic equipment and storage medium | |
| CN110619535A (en) | Data processing method and device | |
| CN110569435B (en) | Intelligent dual-ended recommendation engine system and method | |
| CN116992052B (en) | Long text abstracting method and device for threat information field and electronic equipment | |
| CN105787004A (en) | Text classification method and device | |
| CN111353728A (en) | Risk analysis method and system | |
| CN115544235A (en) | Power grid planning intelligent question-answering system based on text parsing | |
| CN115858939A (en) | Method, system and storage medium for recalling in-line | |
| CN112949305B (en) | Negative feedback information acquisition method, device, equipment and storage medium | |
| KR101178998B1 (en) | Method and System for Certificating Data | |
| CN115482075A (en) | Financial data anomaly analysis method and device, electronic equipment and storage medium | |
| CN109919811B (en) | Insurance agent culture scheme generation method based on big data and related equipment | |
| CN113886547A (en) | Client real-time conversation switching method and device based on artificial intelligence and electronic equipment | |
| CN106933848A (en) | A kind of method for sending information and device | |
| CN111488327A (en) | Data standard management method and system |
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 | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20230816 Address after: 215000 No.35 Shishan Road, high tech Zone, Suzhou City, Jiangsu Province Patentee after: Suzhou Huayi Business Co.,Ltd. Address before: Room 1701, Building 1, No. 35 Shishan Road, High tech Zone, Suzhou City, Jiangsu Province, 215000 Patentee before: Suzhou Huace Network Technology Co.,Ltd. |