Felipe Aguilera

Topic-based social network analysis for virtual communities of interests in the Dark Web

The study of extremist groups and their interaction is a crucial task in order to maintain homeland security and peace. Tools such as social networks analysis and text mining have contributed to the understanding of this kind of groups in order to develop counter-terrorism applications. This work addresses the topic-based community key members extraction problem, for which our method combines both text mining and social network analysis techniques. This is achieved by first applying latent Dirichlet allocation to build two topic-based social networks: one social network oriented towards the thread creator point-of-view, and the other one oriented towards the repliers of the overall forum. Then, by using different Social Network Analysis measures, topic-based key members are evaluated using as benchmark a social network built using the plain documents. Experiments were performed using an English language based forum available in the Dark Web portal.

Virtual Communities of Practice's Purpose Evolution Analysis Using a Concept-Based Mining Approach

Today, social networks systems have become more and more important. People have change their way to relate and communicate. Therefore, how to enhance contents and organization of a social network is a very important task. This way, we can help Virtual communities of practice (VCoP) to survive through time. VCoP are special kind of social network where the purpose is a key aspect. However, administrators are blind when trying to identify how to enhance the community. We propose a method which helps them by analyzing how purpose evolves through time. The approach has been experimentally tested in a real site with successful results.

Topic-based social network analysis for virtual communities of interests in the dark web

The study of extremist groups and their interaction is a crucial task in order to maintain homeland security and peace. Tools such as social networks analysis and text mining have contributed to their understanding in order to develop counter-terrorism applications. This work addresses the topic-based community key-members extraction problem, for which our method combines both text mining and social network analysis techniques. This is achieved by first applying latent Dirichlet allocation to build two topic-based social networks in online forums: one social network oriented towards the thread creator point-of-view, and the other is oriented towards the repliers of the overall forum. Then, by using different network analysis measures, topic-based key members are evaluated using as benchmark a social network built a plain representation of the network of posts. Experiments were successfully performed using an English language based forum available in the Dark Web portal.

Melatonin protects the cytochrome P450 system through a novel antioxidant mechanism

Melatonin, an endogenous hormone, is used as an antioxidant drug in doses quite higher than the endogenous circulating levels of this hormone. Hepatic endoplasmic reticulum contains the cytochrome P450 (CYP450) system, which catalyzes one biotransformation pathway of melatonin; this organelle is also one of the main sources of reactive oxygen species in cells. Therefore, we proposed that the antioxidant activity of this hormone may have a biological relevance in the organelle where it is biotransformed. To evaluate this postulate, we used Fe(3+)/ascorbate, an oxygen free radical generating system that leads to lipid peroxidation, loss of protein-thiol content, and activation of UDP-glucuronyltransferase in rat liver microsomes. We found that mM concentrations of melatonin prevented all these oxidative phenomena. We also found that Fe(3+)/ascorbate leads to structural alterations in the CYP450 monooxygenase, the enzyme that binds the substrate in the CYP450 system catalytic cycle, probably through direct oxidation of the protein, and also inhibited p-nitroanisole O-demethylation, a reaction catalyzed by the CYP450 system. Notably, melatonin prevented both phenomena at microM concentrations. We provide evidence suggesting that melatonin may be oxidized by oxygen free radicals. Thus, we postulate that melatonin may be acting as an oxygen free radical scavenger, and Fe(3+)/ascorbate-modified melatonin would be directly protecting the CYP450 system through an additional specific mechanism. Pharmacological relevance of this phenomenon is discussed.

Leveraging Social Network Analysis with Topic Models and the Semantic Web

Social Network Analysis (SNA) and Web Mining (WM) techniques are being applied to study the structures of social networks in order to manage their dynamics and predict their evolution. This paper describes how we used Semantically-Interlinked Online Communities (SIOC) ontology to represent the (latent) semantic relationships between the members of a large community forum (about 2,500), Plexilandia. We extended SIOC, taking advantage of topic-based text mining and developed data mining algorithms that used our SIOC extensions to provide a better understanding of the social dynamics of the members of the Plexilandia community. This new understanding helped us to detect and discover the key members of Plexilandia successfuly.

Web Intelligence on the Social Web

The WWW, has become a fertile land where anyone can transform his ideas into real applications to create new amazing services. Therefore, it was just a matter of time until the massive proliferation of virtual communities, social networks, etc. New social structures have been formed by massive use of new technologies. This way, people can relate to other by interests, experiences or needs. In a scenario where WWW has become more important every day, and people is using more often the web to relate to others, to read news, obtain tickets, etc. The need of well organized web sites has become one of the vital goals of enterprises and organizations. To accomplish such task web mining area was born more than a decade ago. Web mining are techniques that help managers (or web sites’ experts) to extract information from a web sites’ content, link structure or visitors’ browsing behavior. This way, it is possible to enhance a web site, obtain visitors’ interests patterns to create new services, or provide very specific adds depending on the navigation preferences of visitors (recommendations systems). In the beginning of the Web, web sites were formed by static pages, this means contents were created usually by the owner of the web sites, or the web masters. These contents usually did not change very much through time since it required effort from administrators. Today, a new paradigm arose, we have a participative Web. The web has evolved to the point that it is composed by dynamic contents created by millions of users collaborating one to each other. Sites like, youtube, Blogger, Twitter, facebook, orkut, flickr, among many other, are part of the social web sites’ phenomenon. For example, twitter had 475,000 members by Feb. 2008 while it had 7,038,000 members by Feb. 2009, which means 1382% of growth. Facebook on the same dates passed from 20,043,000 members to 65,704,000 members which means 228%. The use of web intelligence techniques to explode data stored in these social web has become a natural approach to obtain knowledge from them. Since volumes of data are huge, the use of web intelligence techniques was the natural approach to obtain knowledge from social web sites. However, to study members of a social web site is not only to study a group of people accessing a web site and working together; they establish social relationships through the use of Internet tools allowing the formation shared identity and a shared sense of the world. In order to provide truly valuable information to help managers, web masters and to provide better members’ experience when using the social web site, it is necessary to take into account datas’ social nature in web mining techniques. This chapter focuses on the application web intelligence techniques in combination to social network analysis to study of social web sites. In order to provide truly valuable informaton from social web sites that support a social entity. We show that new techniques need to be focused on the study of underlying social aspects of those social entities to really exploit the datas’ social nature and provide a better understandig of human relationships.

A cognitive model of user interaction as a guideline for designing novel interfaces

Adaptive systems behavior based on user models appear promising, mostly for complex environments such as mixed reality environments (MRE). An MRE comprises a virtual representation of the reality as well as physical objects augmented with virtual features. These objects are coupled with the virtual representation so that they can reflect its changes in real time. The proper design of an MRE and the user models that it implies are crucial for its success, but unfortunately, there are no guidelines for the design of these environments. In this paper we present a methodology for designing user models for MRE as well as for the augmentation of physical everyday objects. The user model describes users’ knowledge in two levels of abstraction: objects manipulation (syntax) and its meaning assigned by a community of practice (semantics).

Designing collaborative virtual environments based on real spaces to promote community interaction

A collaborative virtual environment (CVE) is a metaphor of a real environment, but it is not a copy of it. Members of a community may not know each other in real life. The design of CVE in which members are known and there is interaction in a real space is different to the traditional design of CVE. It should consider the real location of each resource, appropriated awareness and communication strategies, and the human-human and human-resource relations. Our university department was selected as an example organizational unit for experimentation. We start with the real physical environment and we design a CVE to provide new collaboration features to people working in the unit and those who will visit it. The advantages of the approach are many. First, people are familiar with the basic physical environment. Second, some activities requiring physical presence can be done with virtual presence, enabling more convenient ways to work for employees. Third, new opportunities for collaborative work appear as it is easy to do them with the proposed CVE. Finally, the approach is extensible, since new features can be added. We present the approach and the design of the proposed CVE.

Leveraging social network analysis with topic models and the Semantic Web extended

Research in the Semantic Web, especially in modeling virtual communities, has provided models useful to represent the richness of these social network interactions. The SIOC Semantically-Interlinked Online Communities vocabulary provides concepts and properties that can be used to describe information from online communities e.g., message boards, wikis, weblogs, etc.. However, the SIOC ontology does not consider social aspects nor the higher order semantics hidden in linkages between community members. This paper describes SIOC-SNA-DM, an extension of the SIOC vocabulary. SIOC-SNA-DMs model is tri-partite, consisting of People, Policies, and Purposes which are social aspects observable in most social communities. A challenge to using our model is how to populate these aspects, since higher order semantics from interactions need to be extracted. Thus, we explain how this population is done with advanced text mining using a latent semantic technique over a large virtual community called Plexilandia.cl with more than 2500 musicians working on the site.Our previous work, in this area, has shown how including these social aspects help to outperform results generated by state-of-the-art techniques. One of the novelties of this present work is the introduction and the elucidation of SIOC-SNA-DM, and how to populate the ontology in order to support the social aspects needed to enhance results of Social Network Mining techniques.

A dissimilarity measure for automate moderation in online social networks

Social networks (SN) have sprouted on the Internet in a very quick way in the last few years. As a large quantity of users starts using them, a lot of social problems are starting to appear, and therefore these sites need to be moderated. However, the data and information volume are so large that it is impossible for a human administrator to handle many of the most common moderation tasks. Web Usage Mining is very useful for understanding user behavior on Websites, opening an opportunity for finding patterns, which can help with decisions afterwards. One of these techniques is clustering, which uses the notion of distance between two behaviors, and tries to capture it among special characteristics. Dissimilarity measures are constructed using important aspects of Website user behavior, but none commonly used ones, such as Cooley et al. distance [3]; help deal with social networking user behavior for moderation tasks. In this work a new dissimilarity measure is used combining usage and contents semantics while interacting with social network platform objects, such as actions, action content, and classification chosen by the user. The measure of this work was successfully tested in a virtual community of practice, obtaining an automatic classification for supporting moderation activities.