May 2016 - Dm-meeting - mailman.cs.vt.edu

Updated summer schedule
by Liangzhe Chen 25 May '16

25 May '16

Hi everyone, unfortunately the meeting room was booked by another group in our regular meeting time, here is what we will do for the summer: -For today we will meet at 2-3pm. -For the rest of the meeting, it will be held on *Thursday *at our regular time (12-1pm). The following is our updated schedule for the summer, and the meeting room has been booked accordingly. -Today at 2pm, Yao -Jun 23, Bijaya -Jun 30, Sorour -Jul 7, Bijaya -Jul 14, Yao Sorry for the inconvenience. Regards, Liangzhe

1 0

DM-Meeting this week
by Liangzhe Chen 24 May '16

24 May '16

Please join us for our next DM-Meeting. ===DM-Meeting Summer 2016=== WHO: Yao Zhang WHEN: Wednesday, May 25, 2016 @ 12:00 noon WHERE: Torg 3160 A WHAT: I will talk about the following two papers: Representation Learning for Information Diffusion through Social Networks: an Embedded Cascade Model, by Simon Bourigault, Sylvain Lamprier, Patrick Gallinari. WSDM 2016. http://dl.acm.org/citation.cfm?id=2835817 Scalable text and link analysis with mixed-topic link models, by Yaojia Zhu, Xiaoran Yan, Lise Getoor and Cristopher Moore. KDD 2013. http://dl.acm.org/citation.cfm?id=2487693 MORE INFO: Here are the brief abstracts of the two papers: Representation Learning for Information Diffusion through Social Networks: an Embedded Cascade Model, WSDM 2016. In this paper, we focus on information diffusion through social networks. Based on the well-known Independent Cascade model, we embed users of the social network in a latent space to extract more robust diffusion probabilities than those defined by classical graphical learning approaches. Better generalization abilities provided by the use of such a projection space allows our approach to present good performances on various real-world datasets, for both diffusion prediction and influence relationships inference tasks. Additionally, the use of a projection space enables our model to deal with larger social networks. Scalable text and link analysis with mixed-topic link models, KDD 2013. Many data sets contain rich information about objects, as well as pairwise relations between them. For instance, in networks of websites, scientific papers, and other documents, each node has content consisting of a collection of words, as well as hyperlinks or citations to other nodes. In order to perform inference on such data sets, and make predictions and recommendations, it is useful to have models that are able to capture the processes which generate the text at each node and the links between them. In this paper, we combine classic ideas in topic modeling with a variant of the mixed-membership block model recently developed in the statistical physics community. The resulting model has the advantage that its parameters, including the mixture of topics of each document and the resulting overlapping communities, can be inferred with a simple and scalable expectation-maximization algorithm. We test our model on three data sets, performing unsupervised topic classification and link prediction. For both tasks, our model outperforms several existing state-of-the-art methods, achieving higher accuracy with significantly less computation, analyzing a data set with 1.3 million words and 44 thousand links in a few minutes. Regards, Liangzhe

1 0

DM-Meeting tomorrow
by Liangzhe Chen 19 May '16

19 May '16

Sorry for the last minute notice, our meeting this week will be on Friday (tomorrow) at the same time (12-1pm). During the summer, we will also continue our meeting regularly. ===DM-Meeting Summer 2016=== WHO: Liangzhe Chen WHEN: Friday, May 20, 2016 @ 12:00 noon WHERE: Torg 3160 A WHAT: I will talk about the following two papers: Modeling Topic Diffusion in Multi-Relational Bibliographic Information Networks <http://hanj.cs.illinois.edu/pdf/cikm14_hgui.pdf> by H. Gui, Y. Sun, J. Han, G. Brova, CIKM'14 Online Influence Maximization <http://delivery.acm.org/10.1145/2790000/2783271/p645-lei.pdf?ip=45.3.66.224…> by S. Lei, S. Maniu, L. Mo, R. Cheng, P. Senellart, KDD'15 MORE INFO: Here are the brief abstracts of the two papers: Modeling Topic Diffusion in Multi-Relational Bibliographic Information Networks <http://hanj.cs.illinois.edu/pdf/cikm14_hgui.pdf> by H. Gui, Y. Sun, J. Han, G. Brova, CIKM'14 Information diffusion has been widely studied in networks, aim- ing to model the spread of information among objects when they are connected with each other. Most of the current research as- sumes the underlying network is homogeneous, i.e., objects are of the same type and they are connected by links with the same se- mantic meanings. However, in the real word, objects are connected via different types of relationships, forming multi-relational hetero- geneous information networks. In this paper, we propose to model information diffusion in such multi-relational networks, by distinguishing the power in passing information around for different types of relationships. We propose two variations of the linear threshold model for multi-relational net- works, by considering the aggregation of information at either the model level or the relation level. In addition, we use real diffu- sion action logs to learn the parameters in these models, which will benefit diffusion prediction in real networks. We apply our diffu- sion models in two real bibliographic information networks, DBLP network and APS network, and experimentally demonstrate the ef- fectiveness of our models compared with single-relational diffusion models. Moreover, our models can determine the diffusion power of each relation type, which helps us understand the diffusion pro- cess better in the multi-relational bibliographic network scenario. Online Influence Maximization <http://delivery.acm.org/10.1145/2790000/2783271/p645-lei.pdf?ip=45.3.66.224…> by S. Lei, S. Maniu, L. Mo, R. Cheng, P. Senellart, KDD'15 Social networks are commonly used for marketing purposes. For example, free samples of a product can be given to a few influential social network users (or seed nodes), with the hope that they will convince their friends to buy it. One way to formalize this objective is through the problem of influence maximization (or IM), whose goal is to find the best seed nodes to activate under a fixed budget, so that the number of people who get influenced in the end is maxi- mized. Solutions to IM rely on the influence probability that a user influences another one. However, this probability information may be unavailable or incomplete. In this paper, we study IM in the absence of complete information on influence probability. We call this problem Online Influence Maximization (OIM), since we learn influence probabilities at the same time we run influence campaigns. To solve OIM, we propose a multiple-trial approach, where (1) some seed nodes are selected based on existing influence information; (2) an influence campaign is started with these seed nodes; and (3) user feedback is used to update influence information. We adopt Explore–Exploit strategies, which can select seed nodes using either the current influence probability estimation (exploit), or the confidence bound on the estimation (explore). Any existing IM algorithm can be used in this framework. We also develop an incremental algorithm that can significantly reduce the overhead of handling user feedback information. Our experiments show that our solution is more effective than traditional IM methods on the partial information. Regards, Liangzhe

1 0

No meeting this week
by Liangzhe Chen 03 May '16

03 May '16

Hi all, we will have no meeting this week due to the upcoming SDM conference that some of us will attend. We will continue our meeting next week though, see you all then! Regards, Liangzhe

1 0