CN103116611A - Social network opinion leader identification method - Google Patents
Social network opinion leader identification method Download PDFInfo
- Publication number
- CN103116611A CN103116611A CN201310028159XA CN201310028159A CN103116611A CN 103116611 A CN103116611 A CN 103116611A CN 201310028159X A CN201310028159X A CN 201310028159XA CN 201310028159 A CN201310028159 A CN 201310028159A CN 103116611 A CN103116611 A CN 103116611A
- Authority
- CN
- China
- Prior art keywords
- node
- seed
- child node
- social networks
- influence power
- 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
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a social network opinion leader identification method which is used for solving the technical problem that the prior social network opinion leader identification method is low in efficiency. The technical scheme is that first, social network data are obtained; then, an information dissemination model is built according to social network characteristics; then, each node degree is calculated according to the information dissemination model and social network topological information; a node which is most influential is used as a first seed node; the left seed nodes in a new node assembly are detected to form a seed node assembly, wherein the seed nodes of the seed node assembly are arranged according to influence from great to small; the first n seed nodes are selected from the seed node assembly with the seed nodes arranged according to influence to constitute an opinion leader assembly, and accordingly, an opinion leader with different influence or charisma is identified. Because the seed node which is most influential in the social network is detected to identify the opinion leader, massive blind detection time is saved and identification efficiency is improved. Compared with the background art, identification efficiency is improved by 50-90 %.
Description
Technical field
The present invention relates to a kind of recognition methods, be specifically related to a kind of social networks leader of opinion recognition methods.
Background technology
Along with the development of Web2.0 technology, social networks (SNS) has become popular network application in the internet.At present, some extensive online social network sites, visit capacity as Facebook has surpassed Google, become the first website of the U.S., and domestic everybody net that is subjected to deeply that the university student welcomes, at present the registered user has reached 100,000,000, and day logs in 2,200 ten thousand person-times (on October 27th, 2009, everybody netted the data of announcement), and the data of alexa in Dec, 2009 website show that in front 15 of domestic and international website visiting amount, social network sites accounts for respectively 4 and 6.Social network sites has millions of online users every day, and this is comprising huge potential business opportunity, promotes their product such as some companies can utilize the social network sites online user.
In social networks, the influence power of planting child node is very important to promoting Information Communication.Some company or users that promote its product, service by virus-type marketing mode nourish very large interest to the influential kind child node of How to choose.Want at social network sites to be that its product is advertised such as A company, because advertising expenditure is limited, can only throw in K user, A company wishes that these initial users can like its product, and with them as kind of a child node, passing on from one to another mode with public praise in social networks affects their friend, allows their friend also like its product, and their friend further affects more friend by social networks, makes more user can like its product.A company wishes that certainly the initial user's (namely planting child node) who selects has considerable influence power, and the number that affects is many as much as possible, just can reach maximum advertisement benefit thereby spend a small amount of expense.As seen, plant child node and brought into play important effect in the spreading network information process, they are equivalent to the leader of opinion, and by their guiding and impact, local suggestion may develop and be network public opinion.The statistics demonstration, the most of user in network does not often participate in manufacturing and the propagation of information, and the leader of opinion is often followed in the decision that they make.Recognition network leader of opinion effectively, delivering guided bone information by the leader of opinion affects the place network user but not directly persuades them, can effectively trigger the influence power of whole network or society, for promoting Information Communication, improve demonstration effect and have important practical significance.
People have studied from different perspectives the social networks leader of opinion and have found and identification problem, identifying the leader of opinion by the kind child node that detects influence power maximum in social networks is wherein a kind of important method, and cause and concern and the attention of industry problems is summed up as the maximizing influence problem.
For the maximizing influence problem, at present derivation algorithm mainly is divided into two classes: (1) complex network algorithm, and such as based on node degree with based on the algorithm at center etc., this class algorithm Main Problems is that resulting kind of child node influence power is on the low side; (2) greedy algorithm, its subject matter be that counting yield is lower, computing time is unstable and extensibility relatively poor etc.
Kempe etc. are at document " Maximizing the spread of influence through a social network(SIGKDD, pages137-146, 2003) ", with the maximizing influence problem as a discrete optimization problems of device, proved that the maximizing influence problem is a NP difficult problem, and an approximate greedy algorithm proposed, its core concept is to choose the maximum node of influence power increment as kind of a child node at every turn, proved that influence power that greedy algorithm obtains kind of child node is not less than (1-1/e) of optimal algorithm, and studied the discrete Information Propagation Model of three kinds of differences and how to have sought the kind child node with maximum effect power in model.Experimental result shows, greedy algorithm obtain kind of child node influence power significantly higher than traditional based on node degree with based on the algorithm at center, but this greedy algorithm has a serious shortcoming, it is exactly the counting yield problem, such as the kind child node of (approximately 15000 nodes) search maximum effect power in social networks on a fairly large scale need to be calculated time a couple of days on a station server, and in extensive social networks, millions of nodes being arranged, may doubly increase by exponentially computing time.
Leskovec etc. are in document " Cost-effective outbreak detection in networks(SIGKDD; pages420-429; 2007) ", propose a CELF (Cost-Effective Lazy Forward selection) and optimize greedy method, the method is based on influence power and has the Submodular function feature extraction, namely the influence power of all nodes weakens along with interstitial content in the seed node set is increased in, and has monotone decline.The method is divided into two steps: first step is used for selecting first kind child node, is all searching for kind of a child node in node, selects the influence power maximum node to join in the seed node set; Second step is used for selecting the remaining child node of planting, and utilizes influence power to have this character of monotone decline and search for kind of a child node in the larger node of some effects power.Due to the minimizing of the search of the method in second step seed node space, so counting yield is greatly improved.The experimental result demonstration, in the situation that Search Results is identical, the CELF algorithm has improved nearly 700 times than the computing velocity of original greedy algorithm, but for large-scale social networks, its counting yield is still lower.
Wei Chen etc. are at document " Efficient influence maximization in social networks (SIGKDD, pp.199-208, 2009) ", a new greedy algorithm (NewGreedy) has been proposed, basic thought is in social network diagram, select dependence edge with factor of influence p between node, set up a brand-new subgraph, then select the subgraph moderate to count the node of maximum as kind of a child node, and a MixGreedy algorithm proposed, it is divided into two parts, first adopts the NewGreedy algorithm idea to choose first kind child node, second portion adopts the CELF algorithm idea to choose the remaining child node of planting.The MixGreedy algorithm combines the advantage of NewGreedy algorithm and CELF algorithm, and its counting yield increases than CELF algorithm.Owing to mutually not activating with factor of influence p between node in linear threshold model, the MixGreedy algorithm need to be tried to achieve kind of a child node from independent cascade model or cum rights cascade model, calculate again their influence power in linear threshold model, therefore its Search Results and other greedy algorithms differ larger sometimes, and extensibility is relatively poor in linear threshold model.
Summary of the invention
In order to overcome the inefficient deficiency of existing social networks leader of opinion recognition methods, the invention provides a kind of social networks leader of opinion recognition methods.the leader of opinion is found and identified to the method by the kind child node that detects influence power maximum in social networks, when planting the child node influence power in detecting social networks, the factors such as relevance according to the distribution of social networks degree of node and degree of node number and influence power, employing detects kind of child node in having the part of nodes of the high number of degrees and it calculates influence power, only need to detect and calculate in the larger node of small part influence power, therefore can save a large amount of blind detection time, reduce kind of a child node influence power computation complexity, improved social networks leader of opinion recognition efficiency.
The technical solution adopted for the present invention to solve the technical problems is: a kind of social networks leader of opinion recognition methods is characterized in comprising the following steps:
(1) utilize the web crawlers instrument, gather actual social network data from the internet.
(2) according to the social networks characteristic, the Information Propagation Models such as the independent cascade model of use carry out modeling analysis to the Information Communication process of social networks.In Information Propagation Model, input social network data and r parameter, wherein r(0＜r≤1) be the number percent that highly several sections of points account for all nodes.
(3) according to information such as Information Propagation Model and social networks topologys, calculate each node number of degrees, and descending the sequence, select the node of the front r of sequence to form new node set.
(4) detect kind of a child node in new node set, and use Submodular function to calculate kind of a child node influence power, with the node of influence power maximum wherein as first kind child node.
(5) detect remaining kind child node in new node set, the same Submodular function that uses calculates kind of a child node influence power, in the process of each selection kind of child node, the large node of calculating section influence power, until all seed node selections are complete, form one and arrive greatly the seed node set of minispread by influence power.
(6) from the seed node set of influence power sequence, choose front n and plant child node, consist of leader of opinion's set, identify the leader of opinion who has Different Effects power or appeal in social networks.In formula, seed interstitial content in 0＜n≤set.
Described r parameter value is between 0.01 to 0.2.
the invention has the beneficial effects as follows: owing to finding and identifying the leader of opinion by the kind child node that detects influence power maximum in social networks, when planting the child node influence power in detecting social networks, the factors such as relevance according to the distribution of social networks degree of node and degree of node number and influence power, employing detects kind of child node in having the part of nodes of the high number of degrees and it calculates influence power, only need to detect and calculate in the larger node of small part influence power, therefore can save a large amount of blind detection time, reduce kind of a child node influence power computation complexity, improved social networks leader of opinion recognition efficiency.Checking and actual test show by experiment, the present invention compares with background technology, in the situation that influence power is not suffered a loss, recognition efficiency has improved 50～90%, and be with good expansibility, be suitable for detecting the kind child node of maximum effect power in extensive social networks, thereby identify the leader of opinion.
Below in conjunction with drawings and Examples, the present invention is elaborated.
Description of drawings
Fig. 1 is the process flow diagram of social networks leader of opinion of the present invention recognition methods.
Embodiment
Key concept involved in the present invention is explained as follows:
(1) Information Propagation Model: social networks is a kind of complex network, and scientific and technological circle adopt Mathematical Modeling Methods that the complex network characteristic is analyzed usually, to describe spreading network information characteristic and inherent law.For social networks, usually adopt three kinds of Information Propagation Models such as independent cascade model, cum rights cascade model and linear threshold model to carry out modeling analysis to social networks kind child node influence power.Like this, how social networks kind child node maximizing influence Solve problems just converts in Information Propagation Model detection and Identification maximum effect power seed node problems to.
(2) influence power function: definition σ () is the influence power function, and S is the seed node set, and U is the search node set.σ (S) is the influence power of seed node set S, and namely S set affects the interstitial content size.
(3) Submodular function character:
Definition 1: if for any element x, y ∈ RK has f (x ∨ y)+f (x ∧ y)≤f (x)+f (y), function f: R
k→ R is Submodular function.
Can be drawn the following conclusions by definition 1.
Conclusion 1: if f is Submodular function,
F (A+j)-f (A) 〉=f (B+j)-f (B) is arranged.As seen, any Submodular function has dullness, the non-character such as negative.
Conclusion 2: in any one example of independent cascade model, cum rights cascade model and linear threshold model, influence power function σ () is a Submodular function.
With reference to Fig. 1.Social networks leader of opinion recognition methods concrete steps of the present invention are as follows:
1. obtain social network data.
Utilize the web crawlers instrument, gather actual social network data from the internet, extracting the network topological informations such as wherein node, connection, to deposit database in pending.
2. set up Information Propagation Model.
According to the social networks characteristic, the Information Propagation Models such as the independent cascade model of use carry out modeling analysis to the Information Communication process of social networks.In Information Propagation Model, the parameters such as input social network data and r, r(0＜r≤1 wherein) account for the number percent of all nodes for highly several sections of points, determine according to parameters such as the interstitial content in social networks, seed interstitial content and network topologies, generally between 0.01 to 0.2.
3. the computing node number of degrees.
According to Information Propagation Model and social networks topology information, calculate each node number of degrees, and descending the sequence, select the node of the front r of sequence to form new node set.
4. select first kind child node.
Detect kind of a child node in new node set, and use Submodular function to calculate kind of a child node influence power, with the node of influence power maximum wherein as first kind child node.
5. select remaining kind child node.
Detect remaining kind child node in new node set, the same Submodular function that uses calculates kind of a child node influence power, in the process of each selection kind of child node, the large node of calculating section influence power, until all seed node selections are complete, form one and arrive greatly the seed node set of little sequence by influence power.
6. identify the leader of opinion.
From the seed node set of influence power sequence, choose seed interstitial content in front n(0＜n≤set) individual kind of child node, consist of leader of opinion's set, thereby identify the leader of opinion who has had Different Effects power or appeal in social networks.
Use false code to describe specific algorithm of the present invention as follows:
Claims (2)
1. social networks leader of opinion recognition methods is characterized in that comprising the following steps:
(1) utilize the web crawlers instrument, gather actual social network data from the internet;
(2) according to the social networks characteristic, the Information Propagation Models such as the independent cascade model of use carry out modeling analysis to the Information Communication process of social networks; In Information Propagation Model, input social network data and r parameter, wherein r(0＜r≤1) be the number percent that highly several sections of points account for all nodes;
(3) according to information such as Information Propagation Model and social networks topologys, calculate each node number of degrees, and descending the sequence, select the node of the front r of sequence to form new node set;
(4) detect kind of a child node in new node set, and use Submodular function to calculate kind of a child node influence power, with the node of influence power maximum wherein as first kind child node;
(5) detect remaining kind child node in new node set, the same Submodular function that uses calculates kind of a child node influence power, in the process of each selection kind of child node, the large node of calculating section influence power, until all seed node selections are complete, form one and arrive greatly the seed node set of minispread by influence power;
(6) from the seed node set of influence power sequence, choose front n and plant child node, consist of leader of opinion's set, identify the leader of opinion who has Different Effects power or appeal in social networks; In formula, seed interstitial content in 0＜n≤set.
2. social networks leader of opinion according to claim 1 recognition methods is characterized in that: described r parameter value is between 0.01 to 0.2.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310028159XA CN103116611A (en) | 2013-01-25 | 2013-01-25 | Social network opinion leader identification method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310028159XA CN103116611A (en) | 2013-01-25 | 2013-01-25 | Social network opinion leader identification method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103116611A true CN103116611A (en) | 2013-05-22 |
Family
ID=48414985
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310028159XA Pending CN103116611A (en) | 2013-01-25 | 2013-01-25 | Social network opinion leader identification method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103116611A (en) |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104134159A (en) * | 2014-08-04 | 2014-11-05 | 中国科学院软件研究所 | Method for predicting maximum information spreading range on basis of random model |
| CN104616200A (en) * | 2015-02-11 | 2015-05-13 | 中南大学 | Influence maximization initial node selecting method based on node characteristics |
| CN105138580A (en) * | 2015-07-31 | 2015-12-09 | 中国科学院信息工程研究所 | Method for minimizing network negative information influence based on blocked continuous boundary |
| CN105335892A (en) * | 2015-10-30 | 2016-02-17 | 南京邮电大学 | Realization method for discovering important users of social network |
| CN105959368A (en) * | 2016-04-29 | 2016-09-21 | 成都信息工程大学 | Social cloud hot spot resource prediction and disposition method |
| CN106204297A (en) * | 2016-07-11 | 2016-12-07 | 深圳市中北明夷科技有限公司 | A kind of recognition methods closing social propagation opinion leader and device |
| CN107123056A (en) * | 2017-03-03 | 2017-09-01 | 华南理工大学 | A kind of location-based social big data information maximization method |
| CN107507020A (en) * | 2017-07-27 | 2017-12-22 | 上海交通大学 | Obtain the maximized method of Internet communication influence power competitive advantage |
| CN108122168A (en) * | 2016-11-28 | 2018-06-05 | 中国科学技术大学先进技术研究院 | Seed node screening technique and device in social activity network |
| CN108710618A (en) * | 2017-04-05 | 2018-10-26 | 中国科学院声学研究所 | One kind is based on concurrent process and the desired propagation seed node selection method of influence power |
| CN110750721A (en) * | 2019-10-21 | 2020-02-04 | 秒针信息技术有限公司 | Information pushing method and device, electronic equipment and readable storage medium |
| CN111125453A (en) * | 2019-12-27 | 2020-05-08 | 中国电子科技集团公司信息科学研究院 | Opinion leader role identification method in social network based on subgraph isomorphism and storage medium |
| CN112100241A (en) * | 2020-09-14 | 2020-12-18 | 哈尔滨工程大学 | Social network dynamic influence maximization method based on theme |
| CN112667876A (en) * | 2020-12-24 | 2021-04-16 | 湖北第二师范学院 | Opinion leader group identification method based on PSOTVCF-Kmeans algorithm |
| US11468521B2 (en) | 2016-10-31 | 2022-10-11 | Tencent Technology (Shenzhen) Company Limited | Social media account filtering method and apparatus |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101859315A (en) * | 2010-04-30 | 2010-10-13 | 西北工业大学 | Heuristic solving method for maximizing influence of social network |
| CN102663101A (en) * | 2012-04-13 | 2012-09-12 | 北京交通大学 | Sina microblog-based user grade sequencing algorithm |
| CN102890696A (en) * | 2011-02-10 | 2013-01-23 | 微软公司 | Social network based contextual ranking |
-
2013
- 2013-01-25 CN CN201310028159XA patent/CN103116611A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101859315A (en) * | 2010-04-30 | 2010-10-13 | 西北工业大学 | Heuristic solving method for maximizing influence of social network |
| CN102890696A (en) * | 2011-02-10 | 2013-01-23 | 微软公司 | Social network based contextual ranking |
| CN102663101A (en) * | 2012-04-13 | 2012-09-12 | 北京交通大学 | Sina microblog-based user grade sequencing algorithm |
Cited By (24)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104134159A (en) * | 2014-08-04 | 2014-11-05 | 中国科学院软件研究所 | Method for predicting maximum information spreading range on basis of random model |
| CN104134159B (en) * | 2014-08-04 | 2017-10-24 | 中国科学院软件研究所 | A kind of method that spread scope is maximized based on stochastic model information of forecasting |
| CN104616200A (en) * | 2015-02-11 | 2015-05-13 | 中南大学 | Influence maximization initial node selecting method based on node characteristics |
| CN104616200B (en) * | 2015-02-11 | 2017-10-10 | 中南大学 | A kind of maximizing influence start node choosing method based on nodal properties |
| CN105138580A (en) * | 2015-07-31 | 2015-12-09 | 中国科学院信息工程研究所 | Method for minimizing network negative information influence based on blocked continuous boundary |
| CN105138580B (en) * | 2015-07-31 | 2018-11-23 | 中国科学院信息工程研究所 | A kind of network negative information influence minimum method based on the company of blocking side |
| CN105335892A (en) * | 2015-10-30 | 2016-02-17 | 南京邮电大学 | Realization method for discovering important users of social network |
| CN105959368A (en) * | 2016-04-29 | 2016-09-21 | 成都信息工程大学 | Social cloud hot spot resource prediction and disposition method |
| CN105959368B (en) * | 2016-04-29 | 2019-04-02 | 成都信息工程大学 | A kind of method of social activity cloud hot point resource prediction and deployment |
| CN106204297A (en) * | 2016-07-11 | 2016-12-07 | 深圳市中北明夷科技有限公司 | A kind of recognition methods closing social propagation opinion leader and device |
| US11468521B2 (en) | 2016-10-31 | 2022-10-11 | Tencent Technology (Shenzhen) Company Limited | Social media account filtering method and apparatus |
| CN108122168A (en) * | 2016-11-28 | 2018-06-05 | 中国科学技术大学先进技术研究院 | Seed node screening technique and device in social activity network |
| CN108122168B (en) * | 2016-11-28 | 2020-11-13 | 中国科学技术大学先进技术研究院 | Method and device for screening seed nodes in social activity network |
| CN107123056A (en) * | 2017-03-03 | 2017-09-01 | 华南理工大学 | A kind of location-based social big data information maximization method |
| CN108710618A (en) * | 2017-04-05 | 2018-10-26 | 中国科学院声学研究所 | One kind is based on concurrent process and the desired propagation seed node selection method of influence power |
| CN107507020B (en) * | 2017-07-27 | 2021-10-08 | 上海交通大学 | Method for obtaining network propagation influence competitive advantage maximization |
| CN107507020A (en) * | 2017-07-27 | 2017-12-22 | 上海交通大学 | Obtain the maximized method of Internet communication influence power competitive advantage |
| CN110750721A (en) * | 2019-10-21 | 2020-02-04 | 秒针信息技术有限公司 | Information pushing method and device, electronic equipment and readable storage medium |
| CN111125453A (en) * | 2019-12-27 | 2020-05-08 | 中国电子科技集团公司信息科学研究院 | Opinion leader role identification method in social network based on subgraph isomorphism and storage medium |
| CN111125453B (en) * | 2019-12-27 | 2023-03-28 | 中国电子科技集团公司信息科学研究院 | Opinion leader role identification method in social network based on subgraph isomorphism and storage medium |
| CN112100241A (en) * | 2020-09-14 | 2020-12-18 | 哈尔滨工程大学 | Social network dynamic influence maximization method based on theme |
| CN112100241B (en) * | 2020-09-14 | 2023-07-25 | 哈尔滨工程大学 | Social network dynamic influence maximization method based on theme |
| CN112667876A (en) * | 2020-12-24 | 2021-04-16 | 湖北第二师范学院 | Opinion leader group identification method based on PSOTVCF-Kmeans algorithm |
| CN112667876B (en) * | 2020-12-24 | 2024-04-09 | 湖北第二师范学院 | Opinion leader group identification method based on PSOTVCF-Kmeans algorithm |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103116611A (en) | Social network opinion leader identification method | |
| CN103678669B (en) | Evaluating system and method for community influence in social network | |
| CN104134159B (en) | A kind of method that spread scope is maximized based on stochastic model information of forecasting | |
| CN103488678B (en) | Friend recommendation system based on user sign-in similarity | |
| CN102394798B (en) | Multi-feature based prediction method of propagation behavior of microblog information and system thereof | |
| CN103678613B (en) | Method and device for calculating influence data | |
| Kundu et al. | A new centrality measure for influence maximization in social networks | |
| CN101957834B (en) | Content recommending method and device based on user characteristics | |
| CN103412872B (en) | Micro-blog social network information recommendation method based on limited node drive | |
| US20170154267A1 (en) | Discovering signature of electronic social networks | |
| Pal et al. | Centrality measures, upper bound, and influence maximization in large scale directed social networks | |
| CN103064917A (en) | Specific-tendency high-influence user group discovering method orienting microblog | |
| CN102262681A (en) | Method for identifying key blog sets in blog information spreading | |
| CN105095419A (en) | Method for maximizing influence of information to specific type of weibo users | |
| CN105550275B (en) | A kind of microblogging transfer amount prediction technique | |
| CN103136331A (en) | Micro blog network opinion leader identification method | |
| CN107123055A (en) | A kind of social big data information maximization method based on PageRank | |
| CN105678590A (en) | topN recommendation method for social network based on cloud model | |
| CN109783805A (en) | A kind of network community user recognition methods and device | |
| Li et al. | Social network user influence dynamics prediction | |
| CN105335476A (en) | Method and device for classifying hot event | |
| Wang et al. | A study on influential user identification in online social networks | |
| CN102750288B (en) | A kind of internet content recommend method and device | |
| CN105589916A (en) | Method for extracting explicit and implicit interest knowledge | |
| CN103200073B (en) | A kind of information processing method and device |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20130522 |