CN103995886B - A kind of various dimensions product-design knowledge pushes framework and construction method - Google Patents
A kind of various dimensions product-design knowledge pushes framework and construction method Download PDFInfo
- Publication number
- CN103995886B CN103995886B CN201410239135.3A CN201410239135A CN103995886B CN 103995886 B CN103995886 B CN 103995886B CN 201410239135 A CN201410239135 A CN 201410239135A CN 103995886 B CN103995886 B CN 103995886B
- Authority
- CN
- China
- Prior art keywords
- push
- dimensional characteristics
- knowledge
- various dimensions
- user
- 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
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
Abstract
The present invention proposes that a kind of various dimensions product-design knowledge pushes framework and construction method.Step 1, according to enterprise product design requirement set up by knowledge push centered on dimensional characteristics and dimensional characteristics analogous relationship table；Step 2, according to dimensional characteristics analogous relationship table set up the hybrid subscriber model based on various dimensions feature；Step 3, tentatively filtered using the behavioral data set pair knowledge of user；Step 4, after the opportunity that triggering is pushed is met, Knowledge delivery after the preliminary filtering that step 3 is obtained gives the sub- engine of push of each dimensional characteristics, and the hybrid subscriber model set up using step 2 calculates the push result of each dimensional characteristics respectively；Step 5, the push result that step 4 is obtained is merged after show active user；If step 6, user's triggering push End Event, this pushes computing and stopped；User terminates knowledge and pushed before use, whenever there is behavior generation, then repeat step 3 to 6.
Description
Technical field
The invention belongs to Computer Applied Technology field, and in particular to a kind of various dimensions product-design knowledge push framework with
Construction method.
Background technology
In increasingly fierce market competition, the knowledge-intensive enterprise for being engaged in product design accumulates rapidly during operation
The a large amount of knowledge networks for constituting core competitiveness are tired out.The implementation of information management makes Knowledge Stock gradually increase, and Knowledge Source becomes
In abundant；But during in face of substantial amounts of knowledge, gradually exposed the problem of knowledge needed for user effectively can not immediately obtain.
How designer is effectively directed to, transmission and the Push Service of knowledge is realized, to reach knowledge in enterprise
The shared, recreation of knowledge, it has also become the problem of in the urgent need to address.Different enterprises under same group, and turn of the market
The same enterprise of different times under constantly accelerating, its operation flow may be entirely different, and this causes its structure of knowledge different, uses
The scene of knowledge is also different from demand.Therefore for different external changes, build adaptive various dimensions knowledge and push framework
It is necessary to meet the changeable knowledge requirement of employee.
Currently as the push for the fundamental way for solving to be difficult in current enterprise " knowledge ocean " knowledge problem needed for obtaining
Technology, the research of its each side is especially burning hot.Various dimensions are one of trend of push technology development, are also that various algorithms are weighed very much
The break-through point wanted.But for current, rarely have scholar to carry out the framework pushed towards product design various dimensions knowledge both at home and abroad
Research.Existing rule-based method can bring preferable result, but need user's customization operations excessive, performance difficulty；Collaboration
Filtering technique has higher automaticity, low to structured data request and can bring higher accuracy rate, but can be by
The problems such as openness and cold start-up.
The current research to knowledge push is only limitted to of push process itself, how includes how to represent knowledge and classification
Reduce error of prediction scoring etc..The structure and follow-up whole calculating processes for pushing various dimensions are considered as an entirety
And it is few.In short, the push process of existing technology description is often in the case of existed system daily record, to data
The excavation of various modes is carried out, the subject matter considered during push is how to produce the higher push of precision.Its tangible record system
The user behaviors log recording mode of various dimensions should be set up according to the particular community of supplying system before system user behaviors log, is then examined again
Consider and how to produce push with accurate algorithm.Existing push technology defect, which is that, to fail in advance to consider the factor of various dimensions
And set up into supplying system behavior record so that accurate and complexity push algorithm is run in coarse data, causes to push away
The imbalance of algorithm and data is sent, the performance of the advantage of knowledge supplying system is influenceed.
Existing method for pushing and theoretical not enough improving will cause：
1) statistical calculation unification.Under single dimension, behavioral data causes quantity pole due to not carrying out various dimensions decomposition
To be huge and there is more noise data.Even if using Cluster Classification algorithm, also simply carry out estimating in data source stealthyly,
Make computing numerous and diverse time-consuming；Single due to originating, the result finally pushed fails to consider the various dimensions of knowledge requirement during product design
Characteristic, causes effect not good.
2) openness and cold start-up.It is openness with cold start-up problem that common influence pushes having for quality.Under conventional frame,
The behavior of user can seem very sparse, and the effect of collaborative filtering scheduling algorithm is substantially reduced, referred to as sparse sex chromosome mosaicism.And newly enter
User and knowledge because behavior is less and is difficult to be associated, cause these users to be difficult to receive the knowledge effectively pushed, this
A little knowledge also are difficult to be pushed to user, and this phenomenon is exactly cold start-up.It is openness to be easily caused push matter with cold start-up problem
Amount reduction, makes supplying system enter vicious circle.
3) various dimensions method is not perfect enough.There is scholar to propose that the knowledge of various dimensions was pushed in the past, but not from dimension
Set up system of selection and propose a set of framework completely and method to the fusion of last various dimensions, so that being difficult to really implement multidimensional
The knowledge of degree is pushed.
4) it is difficult to routinely improve.The information of traditional user behaviors log record is very few, the meaning of behavior behind be difficult to by
Effectively excavate, or even cause wrong deciphering.According to " useless input, useless output " principle (GIGO), the multidimensional of behavior is not considered
Degree characteristic easily causes invalid improvement, and influence pushes algorithm and adaptively lifted.
The content of the invention
The invention aims to overcome the defect of prior art, Knowledge Capability deficiency is obtained in order to solve designer
The problem of, propose that a kind of various dimensions product-design knowledge pushes framework and construction method.
The inventive method is achieved through the following technical solutions：
A kind of various dimensions product-design knowledge pushes framework and construction method, comprises the following steps：
Step 1, according to enterprise product design requirement set up by knowledge push centered on dimensional characteristics and dimensional characteristics phase
Like contingency table；
Step 2, according to dimensional characteristics analogous relationship table set up the hybrid subscriber model based on various dimensions feature；
Step 3, tentatively filtered using the behavioral data set pair knowledge of user；
Step 4, after the opportunity that triggering is pushed is met, Knowledge delivery after the preliminary filtering that step 3 is obtained gives each
The sub- engine of push of dimensional characteristics, the hybrid subscriber model set up using step 2 calculates the push knot of each dimensional characteristics respectively
Really；
Step 5, the push result that step 4 is obtained is merged after show active user；
If step 6, user's triggering push End Event, this pushes computing and stopped；User, which terminates a knowledge push, to be made
With preceding, whenever there is behavior generation, then repeat step 3 to 6；
Since then, the process that various dimensions knowledge pushes framework and system constructing is just completed.
The innovative point of the present invention and the effect reached are mainly reflected in the following aspects：
(1) foundation of various dimensions makes push have more foundations when calculating, and enhances the disposal ability to noise data,
Improve the quality of user model；
(2) it is easy to provide the explanation of push, makes push transparent procedures, the content and inspection pushed is used beneficial to user
System problem；
(3) it can preferably overcome openness, to push to calculate due to the deciphering of various dimensions when data are a small amount of and do not tie
The situation of fruit is greatly decreased；
(4) cold start-up problem is preferably overcome, the user newly added is only that user this dimension data is less, but very
There is the situation that multiple dimensions are all new data simultaneously in hardly possible；The knowledge newly added will have no small machine due to the presence for dimension of classifying
It can be pushed to and be ready the user for receiving new knowledge, so that this framework overcomes the cold start-up of single dimension presence from itself to a certain degree
Problem；
(5) a set of perfect various dimensions method is proposed so that various dimensions are practical, really given play to multiple dimensions
Advantage；
(6) push calculating hour operation quantity small, and push content real-time update, possess excellent novelty；
(7) knowledge requirement of user's various dimensions can be adaptively met by study, enhancing enterprise staff obtains knowledge
Ability.
Whole framework can instruct the building process of enterprise's supplying system when carrying out information management with construction method, so that
Strengthen the Knowledge delivery and acquisition capability of employee, it is ensured that advantage of the enterprise in market competition.
Brief description of the drawings
Fig. 1 is that various dimensions product-design knowledge of the present invention pushes framework and construction method flow chart；
Fig. 2 is the vivid schematic diagram of knowledge push result displaying in the specific embodiment of the invention.
Embodiment
The present invention is described in detail with reference to the accompanying drawings and detailed description.
A kind of various dimensions product-design knowledge of the present invention pushes framework and construction method, specifically real according to following steps
Apply：
Step 1, according to enterprise product design requirement comb and set up by knowledge push centered on dimensional characteristics and dimension
Feature analogous relationship table.
It is the specific object or things being related in whole knowledge supplying system to define 1 dimension, mainly have user, role,
Task, design object, the time, place, use software, knowledge etc..Defining dimension is
Dimension=user, role, task, design object, the time, place, using software, knowledge ...
It is to be under the jurisdiction of dimension and the attribute for characterizing certain dimension to define 2 dimensional characteristics.Dimensional characteristics are specific to dimension
There may be one or more dimensional characteristics under the extension of change, each dimension.The main method of the system is based on dimensional characteristics exhibition
Open.There is main dimensional characteristics in this framework：Individual subscriber, user role, design objective, design objective classification, design object,
Each attribute of design object, time of the act, place, use a variety of classification of software assembly, knowledge, knowledge etc..
Dimensional characteristics={ individual subscriber, user role, design objective, design objective classification, design object, design object
Attribute 1, design object attribute 2 ..., design object attribute N, time of the act, place uses software assembly, knowledge, knowledge classification
1, knowledge classification ..., knowledge classification M etc.
Step 1.1, comb and determine by knowledge push centered on multiple dimensions.Can be according to enterprise's real needs to silent
Recognize dimension content to be adjusted.
Step 1.2, the dimensional characteristics related to knowledge push are combed out according to dimension.Can be according to enterprise's real needs pair
Default dimensions feature is adjusted.
Step 1.3, according to dimensional characteristics set up by knowledge push centered on dimensional characteristics analogous relationship table, each dimension
Feature could set up the similar table on other a certain features in eigen, and the incidence relation of the two features is recorded in this table
In, mainly express the necessity that similarity is set up between each dimensional characteristics, and similarity sets up mode, is empty then not similar
Incidence relation.Such as the similarity relation of row " design object " and row " design object attribute 1 (design object purposes is with using scope) "
For " having ", represent that design object possesses one on purposes and the similarity table using scope.Specific each similarity table
Step 2.2 is shown in foundation.
Step 1.4, the dimensional characteristics collection needed according to dimensional characteristics analogous relationship table, selection, are that wherein each dimension is special
Levy and set up various dimensions similar op, formation pushes sub- engine accordingly.Step 1.3 must be met by pushing the similarity that sub- engine uses
Dimensional characteristics analogous relationship table, the mode of Input knowledge must meet the result that step 3 is tentatively filtered, and output, which must meet, to be defined in 6
Push result specification, above is pushing constraint of the framework to sub- engine.Constrain herein it is upper can any structure again as needed
Build or change the sub- engine of push.
It is, for current a certain feature, the fortune pushed to be produced by one of two ways to define 3 various dimensions similar ops
Calculate.1) first its most like N is taken by the dimensional characteristics1Individual value, then takes the N most like with these values again2Individual value, so weight
After multiple X times, the Knowledge Set most strong and newest with these feature associations is enclosed push and explained, push result is used as；2) first obtain
Most strong and newest knowledge is associated with the value of the dimensional characteristics, its most like N bar knowledge is taken, then takes again with these knowledge most
Similar M1Obtained Knowledge Set, after so repeating X times, is enclosed push and explained, be used as push result by individual knowledge.User, spy
Levy, the respective similarity of knowledge can have a variety of, therefore there can be multiple sub- engines of push with the feature of dimension.Pass through A/B
Test dividing technology and user behavior feedback, using artificial neural network learning optimize N, M ..., X value.
Step 2, according to dimensional information set up the hybrid subscriber model based on various dimensions feature.
Step 2.1, whenever recording a vaild act, all dimensional characteristics for being related in this behavior will be produced and recorded
Onto daily record.Step 2.2 is skipped if existing similarity table, step 2.3 is directly entered.
Define the set that 4 behaviors are user's all behavioural informations in supplying system.It is main include collecting, share, paying close attention to,
Scroll bar etc. is dragged, is the most original data source for portraying user model.Then defining behavior is
Behavior={ collection, is commented on, and scoring is shared, and drags scroll bar, and clickthrough clicks on button ... }
It is to meet the behavior for assert the representative condition of certain behavior to define 5 vaild acts.Common condition is such as browsed
Time, comment fraction, dragging scroll bar number of times etc..Then defining vaild act is
Vaild act=A | A ∈ behaviors, attribute (A) ∈ vaild acts condition set }
Step 2.2, the dimensional characteristics analogous relationship table set up according to step 1.3 set up two-dimentional similarity table.It is each similar
The ranks of degree table are the type of the row of dimensional characteristics analogous relationship table (such as the value " vaild act " of the 13rd row first row
Arrange " individual subscriber "), value in similar table for two values of the type similarities.
Step 2.3, renewal similarity.Vaild act log to being unused for updating similarity, if row dimensional characteristics
Met with row dimensional characteristics：When the values of row dimensional characteristics is identical, row dimensional characteristics values are different, then the different value of row dimensional characteristics
Between increase certain similarity.The behavior record generally occurred jointly is more, and the popularity of the value of ranks type is lower, similar
Degree is higher, can or selection different Similarity Measure function suitably modified according to specific dimensional characteristics.Before increase can according to when
Between necessarily decayed.
It step 2.4, will be marked, will not use again later for the vaild act log for having updated each similarity
In renewal.
Step 3, tentatively filtered using the behavioral data set pair knowledge of user.Specifically implement according to following steps：
Step 3.1, the alternative Knowledge Set for determining push；
Clearly mark is not desired to the knowledge pushed by step 3.2, removal wherein user；
Step 3.3, set selection to exclude according to user once to produce the knowledge of behavior in the range of user's certain time；
Step 3.4, set selection to exclude certain attribute according to user to meet/be unsatisfactory for the knowledge of certain condition.
Step 4, meet triggering push opportunity after, the Knowledge delivery that step 3 is obtained is pushed away to each dimensional characteristics
Sub- engine is sent, the push result of each dimensional characteristics is calculated respectively.Specifically implement according to following steps：
Step 4.1, the selected opportunity for needing to push, such as when Preliminary design product, trial-production change product, arbitrarily browsed；
The dimensional characteristics that step 4.2, acquisition are currently in, alternatively collection passes to each to the knowledge after step 3 is filtered
The sub- engine of push of dimensional characteristics, obtains pushing result by various dimensions similar op；
It is to explain one constituted completely by the knowledge content pushed, push level and push to define 6 and push result
Object.The push level of the push result of each of which intermediate computations must be explained with a certain push and associated.Only last exhibition
The push level that the push result shown allows the push of same type to explain is merged.
Define 7 and push explain it is to explain push the brief sentence of knowledge content in this result why to user, such as according to
The user that role's similarity at family is found, its knowledge interpretation for referring to and approving can be written as " the user XXX similar to your role
Once it with reference to the knowledge ".
It is the significance level that a certain bar knowledge is pushed out in terms of a certain push explanation to define 8 push levels.
Step 5, will push result merged after show active user.Specifically implement according to following steps：
Step 5.1, traversal push each in result and push the push level explained, same type of push is solved respectively
The horizontal averaged released；
The average value of the type calculated in step 5.2, level divided by step 5.1 that each push is explained, multiplied by
The upper weight for setting the dimensional characteristics.Weight can be optimized by evaluating feedback learning.
Step 5.3, the push result for drawing each sub- engine are merged according to the ID of knowledge, while plus with each knowledge ID
Under push level draw total push level；
Step 5.4, by always push intercepted after horizontal descending sort before a number of push result tied as final push
Really；
Step 5.5, for finally pushing each push result, by each explanation push level carry out descending sort after intercept
Preceding a number of interpretative reservation, returns to foreground and shows user.Fig. 2 show the knowledge supplying system of the present embodiment offer
Knowledge pushes the vivid schematic diagram that result is finally shown.
If step 6, user's triggering push End Event, this pushes computing and stopped；User, which terminates a knowledge push, to be made
With preceding, whenever there is behavior generation, then repeat step 3 to 6.
The arrangement flow chart of step 2 to 6 is as shown in Figure 1.To save computing resource, step 2 is generally separated with other steps
Open and carry out at one's leisure.
Since then, just complete/realize the process that various dimensions knowledge pushes framework and system constructing.
Claims (4)
1. a kind of various dimensions product-design knowledge pushes framework and construction method, it is characterised in that comprise the following steps：
Step 1, according to enterprise product design requirement set up by knowledge push centered on dimensional characteristics and dimensional characteristics it is similar close
Join table；
Step 2, according to dimensional characteristics analogous relationship table set up the hybrid subscriber model based on various dimensions feature；
Step 3, tentatively filtered using the behavioral data set pair knowledge of user；
Step 4, after the opportunity that triggering is pushed is met, Knowledge delivery after the preliminary filtering that step 3 is obtained gives each dimension
The sub- engine of push of feature, the hybrid subscriber model set up using step 2 calculates the push result of each dimensional characteristics respectively；
Step 5, the push result that step 4 is obtained is merged after show active user；
If step 6, user's triggering push End Event, this pushes computing and stopped；User terminate knowledge push before use,
Whenever there is behavior generation, then repeat step 3 to 6；
Since then, the process that various dimensions knowledge pushes framework and system constructing is just completed.
2. a kind of various dimensions product-design knowledge as claimed in claim 1 pushes framework and construction method, it is characterised in that build
Vertical dimensional characteristics and dimensional characteristics analogous relationship table use following methods：
Step 1.1, determine by knowledge push centered on multiple dimensions, according to enterprise's real needs to default dimensions content carry out
Adjustment；
Step 1.2, the dimensional characteristics related to knowledge push are combed out according to dimension content, according to enterprise's real needs to acquiescence
Dimensional characteristics content is adjusted；
Step 1.3, according to dimensional characteristics set up by knowledge push centered on dimensional characteristics analogous relationship table；
Step 1.4, the dimensional characteristics collection needed according to dimensional characteristics analogous relationship table, selection, are that wherein each dimensional characteristics are built
Vertical various dimensions similar op, formation pushes sub- engine accordingly.
3. a kind of various dimensions product-design knowledge as claimed in claim 1 or 2 pushes framework and construction method, its feature exists
In described hybrid subscriber model of setting up uses following methods：
Step 2.1, whenever recording a vaild act, then will produce all dimensional characteristics for being related in this behavior and recorded
In daily record, step 2.2 is skipped if existing similarity table, step 2.3 is directly entered；
Step 2.2, the two-dimentional similarity table of dimensional characteristics analogous relationship table foundation according to foundation, the ranks of each similarity table are equal
For the type of the row of dimensional characteristics analogous relationship table, value in similar table for two values of the type similarities；
Step 2.3, renewal similarity, a plurality of vaild act log recording to being unused for updating similarity, if row dimensional characteristics
Met with row dimensional characteristics：The element of row dimensional characteristics and row dimensional characteristics is not empty, and there are row dimensional characteristics in record
The different subset of identical, row dimensional characteristics the value of value, then increase certain similarity between the different value of row dimensional characteristics；
It step 2.4, will be marked, will not be used further to later more for the vaild act log for having updated each similarity
Newly.
4. a kind of various dimensions product-design knowledge as claimed in claim 1 or 2 pushes framework and construction method, its feature exists
In described progress shows active user to use following methods after merging：
Step 5.1, traversal push each in result and push the push level explained, push what is explained to same type of respectively
Horizontal averaged；
The average value of the type calculated in step 5.2, level divided by step 5.1 that each push is explained, multiplied by setting
The weight of the fixed dimensional characteristics；
Step 5.3, the push result for drawing each sub- engine are merged according to the ID of knowledge, while plus with each knowledge ID
Push level draws total push level；
Step 5.4, by always push intercepted after horizontal descending sort before a number of push result as final push result；
Step 5.5, for the final each element pushed in result, carry out intercepting after descending sort by the push level of each explanation
Preceding a number of interpretative reservation, returns to foreground and shows user.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410239135.3A CN103995886B (en) | 2014-05-30 | 2014-05-30 | A kind of various dimensions product-design knowledge pushes framework and construction method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410239135.3A CN103995886B (en) | 2014-05-30 | 2014-05-30 | A kind of various dimensions product-design knowledge pushes framework and construction method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103995886A CN103995886A (en) | 2014-08-20 |
| CN103995886B true CN103995886B (en) | 2017-08-04 |
Family
ID=51310051
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410239135.3A Active CN103995886B (en) | 2014-05-30 | 2014-05-30 | A kind of various dimensions product-design knowledge pushes framework and construction method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103995886B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106406889B (en) * | 2016-09-22 | 2019-10-25 | 广州华克力信息科技有限公司 | Heterogeneous applications system for content intelligent conversion engine suitable for mobile client |
| CN107872483B (en) * | 2016-09-27 | 2020-12-04 | 腾讯科技（北京）有限公司 | Media content pushing method, device and system |
| CN106844470B (en) * | 2016-12-22 | 2020-04-24 | 国网福建省电力有限公司 | Graphics display and analysis method of general field model based on SVG display technology |
| CN107194609B (en) * | 2017-06-15 | 2020-10-09 | 北京理工大学 | Product design system and method |
| CN116401468B (en) * | 2023-04-17 | 2023-12-19 | 哈尔滨工业大学 | Intelligent recommendation system and method based on product concept design characteristic combination |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103593336A (en) * | 2013-10-30 | 2014-02-19 | 中国运载火箭技术研究院 | Knowledge pushing system and method based on semantic analysis |
| CN103761289A (en) * | 2014-01-15 | 2014-04-30 | 北京理工大学 | Knowledge pushing method based on historical records |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020049792A1 (en) * | 2000-09-01 | 2002-04-25 | David Wilcox | Conceptual content delivery system, method and computer program product |
-
2014
- 2014-05-30 CN CN201410239135.3A patent/CN103995886B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103593336A (en) * | 2013-10-30 | 2014-02-19 | 中国运载火箭技术研究院 | Knowledge pushing system and method based on semantic analysis |
| CN103761289A (en) * | 2014-01-15 | 2014-04-30 | 北京理工大学 | Knowledge pushing method based on historical records |
Non-Patent Citations (3)
| Title |
|---|
| Bringing knowledge into recommender systems;Luiz F.C. Tomaz等;《Journal of Systems and Software》;20110630;第1-8页 * |
| 基于本体和粗糙集的产品设计知识推送技术;吉祥等;《计算机集成制造系统》;20130131;第19卷(第1期);第392-397页 * |
| 面向产品设计人员的知识推送服务研究;蒋翠清等;《合肥工业大学学报》;20120331;第35卷(第3期);第392-397页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103995886A (en) | 2014-08-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103995886B (en) | A kind of various dimensions product-design knowledge pushes framework and construction method | |
| Saviotti et al. | The co-evolution of innovation, demand and growth | |
| Zhai et al. | Evolutionary analysis of international collaboration network of Chinese scholars in management research | |
| CN106022708A (en) | Method for predicting employee resignation | |
| CN104504138A (en) | Human-based information fusion method and device | |
| CN110009416A (en) | A kind of system based on big data cleaning and AI precision marketing | |
| CN101706926A (en) | Method for investigating and processing cigarette consumption information | |
| CN111126852A (en) | BI application system based on big data modeling | |
| CN104009993A (en) | Trust evaluation method based on fuzzy filtration | |
| Sun et al. | Evaluation method for innovation capability and efficiency of high technology enterprises with interval-valued intuitionistic fuzzy information | |
| CN104765763B (en) | A kind of semantic matching method of the Heterogeneous Spatial Information classification of service based on concept lattice | |
| Yu et al. | Research on promoting factors of “internet plus innovation and entrepreneurship” based on the perspective of cloud analysis and fuzz analysis method | |
| Klisarova-Belcheva et al. | Business intelligence and analytics–contemporary system model | |
| CN104268270A (en) | Map Reduce based method for mining triangles in massive social network data | |
| CN101741666B (en) | Method to realize multi-instances in the workflow by network structure division | |
| Li | Construction of an interactive sharing platform for competitive intelligence data of marine resources under the background of intelligence construction | |
| Wan | Behavior of sports tourism consumers based on cloud computing and mobile big data | |
| Wu et al. | A study of smart construction and information management models of AEC projects in China | |
| CN105373996A (en) | Modeling system based on approval data | |
| CN104299066B (en) | Organizational information method for digging based on Business Process Management system | |
| CN103714488A (en) | Method for optimizing question answering platform in social network | |
| Shuang | Micro-Transmission Mode Analysis of Ideological and Political Education under Big Data Background | |
| Jin | The Impact of Digital Economy on Circulation Efficiency: Empirical Evidence from China's Postal Industry | |
| Zhao | Resource sharing strategy of regional scientific and technological innovation based on big data analysis platform | |
| Liu | A public opinion monitoring system based on big data technology |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |