CN105095408A - Method and apparatus for judging reliability of network expert - Google Patents
Method and apparatus for judging reliability of network expert Download PDFInfo
- Publication number
- CN105095408A CN105095408A CN201510401076.XA CN201510401076A CN105095408A CN 105095408 A CN105095408 A CN 105095408A CN 201510401076 A CN201510401076 A CN 201510401076A CN 105095408 A CN105095408 A CN 105095408A
- Authority
- CN
- China
- Prior art keywords
- expert
- credibility
- influence factor
- unknown
- factor information
- 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.)
- Pending
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
- G06F16/951—Indexing; Web crawling techniques
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a method and an apparatus for judging the reliability of a network expert, wherein the method comprises: acquiring influential factor information of the reliability of a known expert from Internet; determining the weight of each of influential factors based on the influential factor information of the reliability of the known expert; acquiring influential factor information of the reliability of an unknown expert; and determining the reliability of the unknown expert based on the influential factor information of the reliability of the unknown expert and the weight of each of influential factors. In the embodiment of the invention, through calculating the reliability of the unknown expert, more reliable and valuable advice of an expert can be provided for a user.
Description
Technical field
The embodiment of the present invention relates to Internet technical field, particularly relates to a kind of digerait's confidence level decision method and device.
Background technology
Along with developing rapidly of mutual network, people more and more depend on network.When meeting a difficult problem, also relying on internet hunt, expecting that the expert on network provides guiding suggestion.But the level of the expert on network is also all uneven, and the suggestion that some expert provides is also not too credible, there are even some pseudo-experts.So, how the credibility of expert becomes a problem urgently to be resolved hurrily at present in decision network.
At present, the decision method of conventional expert's credibility understands to judge often have different result of determination for different people, have subjectivity, thus make result of determination lack authenticity.
Summary of the invention
The embodiment of the present invention provides a kind of digerait's confidence level decision method and device, can pass through the credibility calculating unknown expert, for user provides more reliable valuable expert opinion.
First aspect, embodiments provides a kind of digerait's confidence level decision method, comprising:
The influence factor information of the credibility of known expert is obtained from internet;
The weight of each influence factor is determined according to the influence factor information of the credibility of described known expert;
Obtain the influence factor information of the credibility of unknown expert;
The credibility of described unknown expert is determined according to the influence factor information of the credibility of described unknown expert and the weight of described each influence factor.
Second aspect, the embodiment of the present invention also provides a kind of digerait's confidence level decision maker, comprising:
First information acquisition module, for obtaining the influence factor information of the credibility of known expert from internet;
Weight determination module, the influence factor information for the credibility according to described known expert determines the weight of each influence factor;
Second data obtaining module, for obtaining the influence factor information of the credibility of unknown expert;
Credibility determination module, determines the credibility of described unknown expert for the influence factor information of the credibility according to described unknown expert and the weight of described each influence factor.
The embodiment of the present invention by obtaining the influence factor information of the credibility of known expert from internet, and determines the weight of each influence factor according to the influence factor information of the credibility of described known expert; After the influence factor information of credibility getting unknown expert, the weight of the described each influence factor determined according to known expert determines the credibility of described unknown expert further, thus provides more reliable valuable expert opinion for user.
Accompanying drawing explanation
The schematic flow sheet of digerait's confidence level decision method that Fig. 1 provides for the embodiment of the present invention one;
The schematic flow sheet of digerait's confidence level decision method that Fig. 2 provides for the embodiment of the present invention two;
The structural representation of digerait's confidence level decision maker that Fig. 3 provides for the embodiment of the present invention three.
Embodiment
Below in conjunction with drawings and Examples, the present invention is described in further detail.Be understandable that, specific embodiment described herein is only for explaining the present invention, but not limitation of the invention.It also should be noted that, for convenience of description, illustrate only part related to the present invention in accompanying drawing but not entire infrastructure.
The executive agent of digerait's confidence level decision method that the embodiment of the present invention provides, can be integrated digerait's confidence level decision maker on the terminal device, this digerait's confidence level decision maker can adopt hardware or software simulating.
Embodiment one
The schematic flow sheet of digerait's confidence level decision method that Fig. 1 provides for the embodiment of the present invention one, as shown in Figure 1, specifically comprises:
S11, obtain the influence factor information of the credibility of known expert from internet;
Wherein, known expert refer to state one's views in internet and through the expert of the official confirmation identity true and false, comprise true expert and pseudo-expert.
S12, determine the weight of each influence factor according to the influence factor information of the credibility of described known expert;
Wherein, the influence factor information of described credibility can set as required, also can distinguish different Expert opinion standards and determine different influence factor information.Specifically can comprise at least one in following influence factor: the article that user comment content, the some amount of praising, the some amount of stepping on, user comment mark, user comment quantity, expert academic title, the degree of recognition of expert in professional domain, expert deliver and books amount.By Learning Algorithm, such as supervised learning algorithm, unsupervised learning algorithm or semi-supervised learning algorithm etc., calculate the weight of above-mentioned each influence factor respectively.
Such as, for true expert, in user comment content in above-mentioned each influence factor good evaluation and the some amount of praising also more, user comment mark, expert academic title and the expert degree of recognition in professional domain is higher, and some the amount of stepping on and user's bad evaluation also lower.For pseudo-expert, the good evaluation in the user comment content in above-mentioned each influence factor and the some amount of praising also fewer, user comment mark, expert academic title and the expert degree of recognition in professional domain is also lower, and some the amount of stepping on also higher.The article delivered due to expert and books amount, all likely forge, and is not therefore very large on the impact of the judgement of true and false expert.Thus can determine that user comment content in each influence factor, the some amount of praising, the some amount of stepping on, user comment mark, user comment quantity, expert academic title and the degree of recognition of expert in professional domain are on judging that the impact of expert is larger, the weight of its correspondence is also higher, and the article that expert delivers and books amount are on judging that the impact of expert is smaller, the weight of its correspondence is also less.
Concrete, the influence factor information of the credibility of a large amount of known expert can be obtained from (such as Baidu's public praise website) internet, comprise the article of the degree of recognition, expert in the expert academic title of true expert and pseudo-expert, specialist field or books, the comment of netizen praises comprising point or put and step on and the information such as comments of netizen, by the learning algorithm (supervised learning, unsupervised learning or semi-supervised learning) in neural network, the weight that each influence factor corresponding to true expert is corresponding can be obtained.
Specifically when realizing, each information factor that affects of each sample can be quantized, and stored in array, arranging expert's true and false identification item in array, in advance identification item corresponding for true expert can be set to 1, identification item corresponding to pseudo-expert is set to 0.Then using the input data of the array of sample as algorithm.
Such as, the influence factor of the known expert obtained comprises user comment content, the point amount of praising, the point amount of stepping on, user comment mark and user comment quantity, the good evaluation of the user comment content that the true expert obtained is corresponding, the point amount of praising, the point amount of stepping on, the article that user comment mark and expert deliver and books amount are respectively 80 after assignment and normalized, 90, 5, 89 and 50, the good evaluation of the user comment content that the pseudo-expert obtained is corresponding, the point amount of praising, the point amount of stepping on, the article that user comment mark and expert deliver and books amount are respectively 5 after assignment process, 0, 95, 5 and 57, above-mentioned two groups of data are inputted in above-mentioned learning algorithm with the form of vector [80909589501] and [50955570], obtain the user comment content that true expert is corresponding, the point amount of praising, the point amount of stepping on, the article that user comment mark and expert deliver and weight corresponding to books amount are respectively 0.9, 0.8,-0.85, 0.93 and 0.1.Show to affect in information factor above-mentioned, the weight of user comment mark is higher by 0.93, can reflect the true and false of expert, and the article that expert delivers and books amount, weight is lower is 0.1, and the impact in reflection expert true or false is the most weak.
In addition, along with the increase of netizen's number of reviews, the weight authenticity of each influence factor obtained is stronger, by continuous machine learning, finally can obtain the true weight for judging unknown expert's true and false.
S13, obtain the influence factor information of the credibility of unknown expert;
S14, determine the credibility of described unknown expert according to the influence factor information of the credibility of described unknown expert and the weight of described each influence factor.
Concrete, the credibility of described unknown expert can adopt score value to represent, such as, 0 ~ 100 point, the higher credibility of score value is stronger; Also grade can be adopted to represent, such as senior expert, average expert and pseudo-expert etc.Such as, the influence factor information of the credibility of the unknown expert obtained comprises the good evaluation in user comment content, the point amount of praising, the point amount of stepping on, the article that user comment mark and expert deliver and books amount are respectively 80 after assignment and normalized, 90, 30, 70 and 90, the weighted user comment content 0.9 that the true expert then obtained in employing above-mentioned steps S12 is corresponding, the point amount of praising 0.8, the point amount of stepping on-0.85, the article that user comment mark 0.93 and expert deliver and books amount 0.1, the mark obtaining the credibility of unknown expert corresponding is 80*0.9+90*0.8+30* (-0.85)+70*0.93+90*0.1=192.6, be 38.52 after normalized, with reference to the rule that following table one is formulated, the Reliability ratio of this unknown expert known is lower, may be pseudo-expert.
The present embodiment by obtaining the influence factor information of the credibility of known expert from internet, and determines the weight of each influence factor according to the influence factor information of the credibility of described known expert; After the influence factor information of credibility getting unknown expert, the weight of the described each influence factor determined according to known expert determines the credibility of described unknown expert further, thus provides more reliable valuable expert opinion for user.
Exemplary, on the basis of above-described embodiment, determine that the weight of each influence factor specifically comprises according to the influence factor information of the credibility of described known expert:
Quantification treatment is carried out to the influence factor information of the credibility of described known expert;
Using the data after the quantification of described known expert as training data, any one algorithm following is adopted to determine the weight of each influence factor: supervised learning algorithm, unsupervised learning algorithm or semi-supervised learning algorithm.
Concrete, when the influence factor information of the credibility to described known expert carries out quantification treatment, based on following principle, the assignment of higher fractional can be given for comment preferably, gives lower assignment for poor comment.
Such as, when the influence factor information of described credibility comprises: when the article that user comment content, the some amount of praising, the some amount of stepping on, user comment mark, user comment quantity, expert academic title, the degree of recognition of expert in professional domain, expert deliver and books amount, for user comment content, when the user comment content got is " this expert opinion is relatively good ", then to the arbitrary value between its assignment 80-90; The article delivered for the some amount of praising, the some amount of stepping on, user comment mark, user comment quantity, expert and books amount, directly can adopt the score value that they show separately; For expert academic title, can assignment be arbitrary value between 80-100 for high title, can assignment be arbitrary value between 60-80 for common academic title, for can assignment be the arbitrary value of less than 60 without academic title; For the degree of recognition of expert in professional domain, for degree of recognition high can assignment be arbitrary value between 80-100, for degree of recognition medium can assignment be arbitrary value between 60-80, for degree of recognition low can assignment be the arbitrary value of less than 60.
Concrete, in supervised learning process, each example is made up of an input object (being generally vector) and an output valve (also referred to as supervisory signals) expected.The input data of supervised learning are called as training data, the influence factor information that the known expert after namely quantizing is corresponding, and use supervised learning algorithm to analyze this training data, and produce a function inferred, it may be used for mapping the example made new advances.The example of those the unknowns and the class label of unknown expert correctly can be determined by supervised learning algorithm.
Specifically can adopt neural network (BackPropagation, BP) algorithm realizes, BP network is conventional supervised learning algorithm, its simulation people brain neuron message processing facility, it is a kind of Multi-layered Feedforward Networks by Back Propagation Algorithm training, can learn and store a large amount of input-output mode map relations, and without the need to disclosing the math equation describing this mapping relations in advance.Its learning rules use method of steepest descent, constantly adjusted the weights and threshold of network, make the error sum of squares of network minimum, thus the weight of each influence factor obtained is tended towards stability by backpropagation.
Exemplary, on the basis of above-described embodiment, determine the credibility of described unknown expert according to the influence factor information of the credibility of described unknown expert and the weight of described each influence factor:
The influence factor information of the credibility of described unknown expert is carried out quantification treatment;
Calculate the weighted value of the weight of the data after the quantification of described unknown expert and described each influence factor;
The credibility of described unknown expert is determined according to described weighted value.
Such as, can using described weighted value directly as the credibility value of described unknown expert, its value is higher, and to represent the credibility of described unknown expert higher.
Expert's table of comparisons can also be set up in advance, preset expert's table of comparisons according to described weighted value inquiry, in described default expert's table of comparisons, comprise the corresponding relation of weighted value and described credibility;
The credibility that described weighted value is corresponding is determined from described default expert's table of comparisons.
Such as, the form as shown in following table one can be set up in advance, in table one, have recorded the credibility that described weighted value is corresponding.When setting up form, usually adopt centesimal system mode, by calculating after the influence factor information quantization of unknown expert, the weighted value of its correspondence normalizes between 0 to 100.
Table one
| Weighted value | Expert's credibility |
| 0–60 | Lower |
| 60—80 | Average expert |
| 80—100 | Authoritative expert |
When described weighted value is between 0-60, then illustrates that the speech credibility of this unknown expert is low, netizen is not helped substantially, pseudo-expert can be thought; When described weighted value is between 60-80, then illustrate that the speech of this unknown expert is substantially credible, helpful to netizen; When described weighted value is between 80-100, then illustrate that the speech of this unknown expert is credible, this unknown expert is authoritative expert, helps greatly netizen.
The various embodiments described above equally by obtaining the influence factor information of the credibility of known expert from internet, and determine the weight of each influence factor according to the influence factor information of the credibility of described known expert; After the influence factor information of credibility getting unknown expert, the weight of the described each influence factor determined according to known expert determines the credibility of described unknown expert further, thus provides more reliable valuable expert opinion for user.
Embodiment two
The schematic flow sheet of digerait's confidence level decision method that Fig. 2 provides for the embodiment of the present invention two, as shown in Figure 2, specifically comprises:
S21, obtain the influence factor information of the credibility of known expert from internet;
Such as, the article that the influence factor information of acquisition comprises user comment content, the some amount of praising, the some amount of stepping on, user comment mark, user comment quantity, expert academic title, the degree of recognition of expert in professional domain, expert deliver and books amount.
S22, quantification treatment is carried out to the influence factor information of the credibility of described known expert;
Such as, assignment process is carried out to each influence factor that above-mentioned steps S21 gets, gives higher assignment for good comment, give lower assignment for bad comment.
S23, using the data after the quantification of described known expert as training data, adopt supervised learning algorithm to determine the weight of each influence factor;
S24, obtain the influence factor information of the credibility of unknown expert;
S25, the influence factor information of the credibility of described unknown expert is carried out quantification treatment;
The process of the influence factor information of the credibility of described unknown expert being carried out to quantification treatment is similar with the process of the influence factor information of the credibility of described known expert being carried out to quantification treatment, specifically see the associated description of in above-described embodiment one, the influence factor information of the credibility of described known expert being carried out to quantification treatment, can repeat no more here.
S26, calculate described unknown expert quantification after data and the weighted value of weight of described each influence factor;
Concrete, if the influence factor of the credibility of the described unknown expert got and above-mentioned known expert's one_to_one corresponding, then ask for the weighted sum of each influence factor of described unknown expert and the weight of described each influence factor, using the value of described weighted sum as described unknown expert.Such as, if be the first score value d1 after the user comment content assignment of the described unknown expert obtained, the some amount of praising is the second score value d2, the some amount of stepping on is the 3rd score value d3, user comment mark is quartile d4, user comment quantity is quintile d5, expert academic title is the 6th score value d6, the degree of recognition of expert in professional domain is the 7th score value d7, expert delivers article and books amount be the 8th score value d8; The article that described user comment content, the some amount of praising, the some amount of stepping on, user comment mark, user comment quantity, expert academic title, the degree of recognition of expert in professional domain, expert deliver and weight corresponding to books amount are respectively n1, n2, n3, n4, n5, n6, n7 and n8, then following formula one finally can be adopted to calculate the weighted value of described unknown expert:
d1n1+d2n2+d3n3+d4n4+d5n5+d6n6+d7n7+d8n8
If the influence factor of the credibility of the described unknown expert got and above-mentioned known expert not one_to_one corresponding, then only ask for the weighted sum of each influence factor corresponding with described known expert, using the value of described weighted sum as described unknown expert.Such as, if the influence factor of the described unknown expert obtained comprises user comment content, the some amount of praising, the some amount of stepping on, user comment mark and expert academic title, then following formula two finally can be adopted to calculate the weighted value of described unknown expert:
d1n1+d2n2+d3n3+d4n4+d6n6
S27, preset expert's table of comparisons according to the inquiry of described weighted value, in described default expert's table of comparisons, comprise the corresponding relation of weighted value and described credibility;
S28, from described default expert's table of comparisons, determine the credibility that described weighted value is corresponding.
Detailed description for step S27 and S28 specifically see the associated description in embodiment one, can repeat no more here.
The present embodiment equally by obtaining the influence factor information of the credibility of known expert from internet, and determines the weight of each influence factor according to the influence factor information of the credibility of described known expert; After the influence factor information of credibility getting unknown expert, the weight of the described each influence factor determined according to known expert determines the credibility of described unknown expert further, thus provides more reliable valuable expert opinion for user.
Embodiment three
The structural representation of digerait's confidence level decision maker that Fig. 3 provides for the embodiment of the present invention three, as shown in Figure 3, specifically comprises: first information acquisition module 31, weight determination module 32, second data obtaining module 33 and credibility determination module 34.
First information acquisition module 31 is for obtaining the influence factor information of the credibility of known expert from internet;
Weight determination module 32 determines the weight of each influence factor for the influence factor information of the credibility according to described known expert;
Second data obtaining module 33 is for obtaining the influence factor information of the credibility of unknown expert;
Credibility determination module 34 determines the credibility of described unknown expert for the influence factor information of the credibility according to described unknown expert and the weight of described each influence factor.
Digerait's confidence level decision maker described in the present embodiment is for performing the digerait's confidence level decision method described in embodiment one to embodiment two, and the technique effect of its know-why and generation is similar, is not repeated here.
Exemplary, on the basis of above-described embodiment, the influence factor information of described credibility comprises at least one in following influence factor:
The article that user comment content, the some amount of praising, the some amount of stepping on, user comment mark, user comment quantity, expert academic title, the degree of recognition of expert in professional domain, expert deliver and books amount.
Exemplary, described weight determination module 32 specifically for:
Quantification treatment is carried out to the influence factor information of the credibility of described known expert;
Using the data after the quantification of described known expert as training data, any one algorithm following is adopted to determine the weight of each influence factor: supervised learning algorithm, unsupervised learning algorithm or semi-supervised learning algorithm.
Exemplary, described credibility determination module 34 specifically comprises:
Quantification treatment unit 341 carries out quantification treatment for the influence factor information of the credibility by described unknown expert;
Weighted value computing unit 342 for calculate described unknown expert quantification after data and the weighted value of weight of described each influence factor;
Credibility determining unit 343 is for determining the credibility of described unknown expert according to described weighted value.
Exemplary, described credibility determining unit 343 specifically for:
Preset expert's table of comparisons according to described weighted value inquiry, in described default expert's table of comparisons, comprise the corresponding relation of weighted value and described credibility;
The credibility that described weighted value is corresponding is determined from described default expert's table of comparisons.
Digerait's confidence level decision maker described in the various embodiments described above is equally for performing the digerait's confidence level decision method described in embodiment one to embodiment two, and the technique effect of its know-why and generation is similar, is not repeated here.
Note, above are only preferred embodiment of the present invention and institute's application technology principle.Skilled person in the art will appreciate that and the invention is not restricted to specific embodiment described here, various obvious change can be carried out for a person skilled in the art, readjust and substitute and can not protection scope of the present invention be departed from.Therefore, although be described in further detail invention has been by above embodiment, the present invention is not limited only to above embodiment, when not departing from the present invention's design, can also comprise other Equivalent embodiments more, and scope of the present invention is determined by appended right.
Claims (10)
1. digerait's confidence level decision method, is characterized in that, comprising:
The influence factor information of the credibility of known expert is obtained from internet;
The weight of each influence factor is determined according to the influence factor information of the credibility of described known expert;
Obtain the influence factor information of the credibility of unknown expert;
The credibility of described unknown expert is determined according to the influence factor information of the credibility of described unknown expert and the weight of described each influence factor.
2. method according to claim 1, is characterized in that, the influence factor information of described credibility comprises at least one in following influence factor:
The article that user comment content, the some amount of praising, the some amount of stepping on, user comment mark, user comment quantity, expert academic title, the degree of recognition of expert in professional domain, expert deliver and books amount.
3. method according to claim 1, is characterized in that, determines that the weight of each influence factor comprises according to the influence factor information of the credibility of described known expert:
Quantification treatment is carried out to the influence factor information of the credibility of described known expert;
Using the data after the quantification of described known expert as training data, any one algorithm following is adopted to determine the weight of each influence factor: supervised learning algorithm, unsupervised learning algorithm or semi-supervised learning algorithm.
4. the method according to any one of claims 1 to 3, is characterized in that, determines the credibility of described unknown expert according to the influence factor information of the credibility of described unknown expert and the weight of described each influence factor:
The influence factor information of the credibility of described unknown expert is carried out quantification treatment;
Calculate the weighted value of the weight of the data after the quantification of described unknown expert and described each influence factor;
The credibility of described unknown expert is determined according to described weighted value.
5. method according to claim 4, is characterized in that, determines that the credibility of described unknown expert comprises according to described weighted value:
Preset expert's table of comparisons according to described weighted value inquiry, in described default expert's table of comparisons, comprise the corresponding relation of weighted value and described credibility;
The credibility that described weighted value is corresponding is determined from described default expert's table of comparisons.
6. digerait's confidence level decision maker, is characterized in that, comprising:
First information acquisition module, for obtaining the influence factor information of the credibility of known expert from internet; Weight determination module, the influence factor information for the credibility according to described known expert determines the weight of each influence factor;
Second data obtaining module, for obtaining the influence factor information of the credibility of unknown expert;
Credibility determination module, determines the credibility of described unknown expert for the influence factor information of the credibility according to described unknown expert and the weight of described each influence factor.
7. device according to claim 6, is characterized in that, the influence factor information of described credibility comprises at least one in following influence factor:
The article that user comment content, the some amount of praising, the some amount of stepping on, user comment mark, user comment quantity, expert academic title, the degree of recognition of expert in professional domain, expert deliver and books amount.
8. device according to claim 6, is characterized in that, described weight determination module specifically for:
Quantification treatment is carried out to the influence factor information of the credibility of described known expert;
Using the data after the quantification of described known expert as training data, any one algorithm following is adopted to determine the weight of each influence factor: supervised learning algorithm, unsupervised learning algorithm or semi-supervised learning algorithm.
9. the device according to any one of claim 6 ~ 8, is characterized in that, described credibility determination module comprises:
Quantification treatment unit, the influence factor information for the credibility by described unknown expert carries out quantification treatment;
Weighted value computing unit, for calculate described unknown expert quantification after data and the weighted value of weight of described each influence factor;
Credibility determining unit, for determining the credibility of described unknown expert according to described weighted value.
10. device according to claim 9, is characterized in that, described credibility determining unit specifically for:
Preset expert's table of comparisons according to described weighted value inquiry, in described default expert's table of comparisons, comprise the corresponding relation of weighted value and described credibility;
The credibility that described weighted value is corresponding is determined from described default expert's table of comparisons.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510401076.XA CN105095408A (en) | 2015-07-09 | 2015-07-09 | Method and apparatus for judging reliability of network expert |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510401076.XA CN105095408A (en) | 2015-07-09 | 2015-07-09 | Method and apparatus for judging reliability of network expert |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105095408A true CN105095408A (en) | 2015-11-25 |
Family
ID=54575845
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510401076.XA Pending CN105095408A (en) | 2015-07-09 | 2015-07-09 | Method and apparatus for judging reliability of network expert |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105095408A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105825570A (en) * | 2016-03-03 | 2016-08-03 | 孙腾 | Network voting display method and system thereof |
| CN106339172A (en) * | 2016-08-25 | 2017-01-18 | 北京小米移动软件有限公司 | Hobby information acquisition method and device |
| CN106815257A (en) * | 2015-12-01 | 2017-06-09 | 阿里巴巴集团控股有限公司 | The computational methods and device of a kind of User reliability |
| CN107491491A (en) * | 2017-07-20 | 2017-12-19 | 西南财经大学 | A kind of media article for adapting to user interest change recommends method |
| CN110413778A (en) * | 2019-07-10 | 2019-11-05 | 深圳传世智慧科技有限公司 | Generation method, expert recommendation method and the electronic equipment of expert power |
| CN111127103A (en) * | 2019-12-24 | 2020-05-08 | 北京阿尔山区块链联盟科技有限公司 | Value evaluation method and system for digital assets |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101399707A (en) * | 2008-11-20 | 2009-04-01 | 北京邮电大学 | Method and device for selecting internet service based on credit |
| CN102456158A (en) * | 2010-10-26 | 2012-05-16 | 中国民航大学 | Security assessment method for air traffic management (ATM) information system based on ANNBP (Artificial Neural Network Blood Pressure) model |
-
2015
- 2015-07-09 CN CN201510401076.XA patent/CN105095408A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101399707A (en) * | 2008-11-20 | 2009-04-01 | 北京邮电大学 | Method and device for selecting internet service based on credit |
| CN102456158A (en) * | 2010-10-26 | 2012-05-16 | 中国民航大学 | Security assessment method for air traffic management (ATM) information system based on ANNBP (Artificial Neural Network Blood Pressure) model |
Non-Patent Citations (1)
| Title |
|---|
| 朱庆锋等: "基于模糊综合评价法和BP神经网络法的企业控制活动评价及比较分析", 《管理评论》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106815257A (en) * | 2015-12-01 | 2017-06-09 | 阿里巴巴集团控股有限公司 | The computational methods and device of a kind of User reliability |
| CN105825570A (en) * | 2016-03-03 | 2016-08-03 | 孙腾 | Network voting display method and system thereof |
| CN106339172A (en) * | 2016-08-25 | 2017-01-18 | 北京小米移动软件有限公司 | Hobby information acquisition method and device |
| CN107491491A (en) * | 2017-07-20 | 2017-12-19 | 西南财经大学 | A kind of media article for adapting to user interest change recommends method |
| CN110413778A (en) * | 2019-07-10 | 2019-11-05 | 深圳传世智慧科技有限公司 | Generation method, expert recommendation method and the electronic equipment of expert power |
| CN111127103A (en) * | 2019-12-24 | 2020-05-08 | 北京阿尔山区块链联盟科技有限公司 | Value evaluation method and system for digital assets |
| CN111127103B (en) * | 2019-12-24 | 2023-10-24 | 北京阿尔山区块链联盟科技有限公司 | Value evaluation method and system for digital asset |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105095408A (en) | Method and apparatus for judging reliability of network expert | |
| Clements et al. | Forecasting economic and financial time-series with non-linear models | |
| Tsinaslanidis et al. | A prediction scheme using perceptually important points and dynamic time warping | |
| Karsoliya | Approximating number of hidden layer neurons in multiple hidden layer BPNN architecture | |
| CN111415492B (en) | Slope landslide early warning method and system based on fuzzy comprehensive evaluation algorithm | |
| CN110232240B (en) | Improved transformer top layer oil temperature prediction method | |
| WO2018028767A1 (en) | Devices and methods for evaluating speech quality | |
| CN110110062A (en) | Machine intelligence answering method, device and electronic equipment | |
| CN113538125A (en) | Risk rating method for optimizing Hopfield neural network based on firefly algorithm | |
| Rady et al. | Time series forecasting using tree based methods | |
| CN111078847A (en) | Power consumer intention identification method and device, computer equipment and storage medium | |
| CN113011895B (en) | Associated account sample screening method, device and equipment and computer storage medium | |
| CN106485585A (en) | Method and system for ranking | |
| US20190362354A1 (en) | Real-time updating of predictive analytics engine | |
| US20170337240A1 (en) | Software function verification system and software function verification method | |
| Majumder et al. | Forecasting of Indian stock market index using artificial neural network | |
| CN111340236A (en) | Bond default prediction method based on bond valuation data and integrated machine learning | |
| CN110796231A (en) | Data processing method, data processing device, computer equipment and storage medium | |
| CN113919437A (en) | Method, device, equipment and storage medium for generating client portrait | |
| Reddy et al. | A concise neural network model for estimating software effort | |
| Sebastian | Performance evaluation by artificial neural network using WEKA | |
| CN114021432A (en) | Stress corrosion cracking crack propagation rate prediction method and system | |
| CN109598410A (en) | Presell methods of risk assessment, system, computer installation and readable storage medium storing program for executing | |
| CN111475631B (en) | Disease question-answering method and device based on knowledge graph and deep learning | |
| CN116739787A (en) | Transaction recommendation method and system based on artificial intelligence |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20151125 |
|
| RJ01 | Rejection of invention patent application after publication |