Ofelia Cervantes

Invited Paper: Intelligent Data Mining Assistance via CBR and Ontologies

Most commercial data mining products provide a large number of models and tools for performing various data mining tasks, but few provide intelligent assistance for addressing many important decisions that must be considered during the mining process. In this paper, we propose the realization of a hybrid data mining assistant, based on the CBR paradigm and the use of an ontology, in order to empower the user during the various phases of the data mining process

Sentiment Groups as Features of a Classification Model Using a Spanish Sentiment Lexicon: A Hybrid Approach

Discovering peoples subjective opinion about a topic of interest has become more relevant with the explosion in the use of social networks, microblogs, forums and e-commerce pages all over the Internet. Sentiment analysis techniques aim to identify polarity of opinions by analyzing explicit and implicit features within the text. This paper presents a hybrid approach to extract features from Spanish sentiment sentences in order to create a model based on support vector machines and determine polarity of opinions. In addition to this, a Spanish Sentiment Lexicon has been constructed. Accuracy of the model is evaluated against two previously tagged corpora and results are discussed.

UDLAP: Sentiment Analysis Using a Graph-Based Representation

We present an approach for tackling the Sentiment Analysis problem in SemEval 2015. The approach is based on the use of a cooccurrence graph to represent existing relationships among terms in a document with the aim of using centrality measures to extract the most representative words that express the sentiment. These words are then used in a supervised learning algorithm as features to obtain the polarity of unknown documents. The best results obtained for the different datasets are: 77.76% for positive, 100% for negative and 68.04% for neutral, showing that the proposed graph-based representation could be a way of extracting terms that are relevant to detect a sentiment.

Visualizing collaboration networks implicit in digital libraries using OntoStarFish

This paper presents the design rationale and initial findings derived from preliminary usage of OntoStarFish, a visualization technique aimed at taking advantage of implicit relationships that can be inferred from large collections of documents in digital libraries. OntoStarFish makes such relationships explicit so users may visualize them and detect potential collaboration networks. Users that may be interested in exploring collaboration networks include researchers looking for partners for specific projects as well as funding agencies concerned with the strength of associations among participants of competing proposals. OntoStarFish is based upon the use of multiple fisheye views that can be placed on top of starfields, dynamic scatter plots for which each axis is determined by a lightweight ontology of attributes associated to potential collaborators.

Author attribution using a graph based representation

Authorship attribution is the task of determining the real author of a given anonymous document. Even though different approaches exist in literature, this problem has been barely dealt with by using document representations that employ graph structures. Actually, most research works in literature handle this problem by employing simple sequences of n words (n-grams), such as bigrams and trigrams. In this paper, an exploration in the use of graphs for representing document sentences is presented. These structures are used for carrying out experiments for solving the problem of Authorship attribution. The experiments that are presented here attain approximately a 79% of accuracy, showing that the graph-based representation could be a way of encapsulating various levels of natural language descriptions in a simple structure.

A Platform for Creating Augmented Reality Content by End Users

We present work in progress towards the development of a platform for the creation of augmented reality (AR) content by the end user. Based upon a review of existing AR authoring tools and scenarios we have envisioned in the context of smart cities, we have developed SituAR, an architecture for a platform in which the user is able to create AR content using multimedia elements. Our emphasis is on making augmented reality easier to put together and to empower users to become authors in AR scenarios. We also include social media elements for users to share, rank, and comment the content created in order to add new information and to facilitate interaction. This paper discusses the architecture of SituAR and its potential.

UDLAP at SemEval-2016 Task 4: Sentiment Quantification Using a Graph Based Representation.

We present an approach for tackling the tweet quantification problem in SemEval 2016. The approach is based on the creation of a cooccurrence graph per sentiment from the training dataset and a graph per topic from the test dataset with the aim of comparing each topic graph against the sentiment graphs and evaluate the similarity between them. A heuristic is applied on those similarities to calculate the percentage of positive and negative texts. The overall result obtained for the test dataset according to the proposed task score (KL divergence) is 0.261, showing that the graph based representation and heuristic could be a way of quantifying the percentage of tweets that are positive and negative in a given set of texts about a topic.

An extensible platform for seamless integration and management of applications for emotion sensing and interpretation

The role of affect has become increasingly important in ambient intelligence applications. Developers require means for using multiple tools that support emotion detection and interpretation, which can be used jointly to provide meaningful system responses for the user in varied situations. We introduce Vikara, an extensible software platform that provides developers with uniform interfaces and services so their applications can access the results from existing (or newly implemented) tools for emotion sensing and interpretation. Vikara also provides components that make it possible for platform managers to visually monitor affective states as they are sensed by the various available tools. We have experimented with our platform via the development of applications for emotion detection by using (1) the well-known Facial Action Coding System and a Kinect sensor; and (2) a self-reporting interface for Android-based mobile devices. We report on the initial results of using our platform and discuss ambient intelligence scenarios in which we see potential applications, including user experience and usability evaluation, marketing, distance education and inter-personal communication.

A pattern-based approach for developing creativity applications supported by surface computing

An activity model for applications focused on creativity.Building blocks for constructing applications designed for creativity.By relying on our building blocks, applications were implemented in short time.Our building blocks prove the models expressivity for describing creativity sessions. Whereas creativity tasks have traditionally been supported by conventional media and tools (such as paper, pens, scissors and glue); mobile phones, tablets and other devices based on interactive surfaces are increasingly been used as additional support. Large-sized multi-touch interactive surfaces appear as an interesting alternative for supporting creativity processes and for supporting synchronous collocated collaboration. However, they have mostly been usedfor visualization and navigation purposes. Their use as authoring means, which would be essential in creativity tasks, has only begun to be explored. Applications and platforms that have been developed in this area rely on low-level primitives for implementing representation of ideas and discussions. We have identified a significant gap between the level of development tools and the abstractions required by end-user applications that aim to support creativity processes using interactive surfaces. This gap makes it difficult for developers to build applications that provide richer, more flexible support for innovators working collaboratively around interactive surfaces. Based upon a thorough analysis of existing applications and user practices in the field, we have identified the key actions and interaction patterns that take place during collaborative creativity sessions. Thus, we propose ISCALI (Innovation Solutions Centered on Activities for Large-sized Interfaces), a model that can be used both for describing and for prescribing the role of multi-touch surfaces in collaborative creativity tasks. In accordance to Activity Centered Design, ISCALI comprises three major components: activities, actions and operations. The central activities within the processes of creativity comprise generation, organization and evaluation of ideas. Each of these activities encompasses sets of actions. Finally, several operation sets achieve the goal of each of the actions. Based upon our model, we designed a general architecture for collaborative creativity applications. This architecture addresses the development gap through a proposed set ofbuilding blocks. These building blocks implement the main interaction patterns needed for stimulating creativity tasks that rely on interactive surfaces. We have implemented prototypical versions of these building blocks, referred to as TOKAs (Touch Operations for Creative Activities), and have made them available to developers. Independent developers have implemented applications that facilitate the use of various methodologies that foster creativity and synchronous collocated collaboration. These developers have taken advantage of the availability of TOKAs. The implementation and use of TOKAs demonstrate ISCALIs expressivity for describing and guiding the development of applications that support collaborative creativity on top of interactive surfaces.

Visualizing sentiment change in social networks

This paper presents SCWorld, a novel scheme for visualizing sentiment changes in real-time in social networks. Two main features distinguish SCWorld from existing work in sentiment classification in real time: First, it provides a high-level, large-granularity view of sentiment and relationships among dynamic clusters of topics in a social network; second, it allows users to observe animated graphical representations of sentiment changes that reflect the aggregated polarity of postings and other user activities in social networks. SCWorld builds upon Expression, a platform for sentiment classification that also provides a low-granularity topic visualization scheme termed Sentiment Card (SC). SCs merge quantitative and qualitative data that arise from sentiment analysis at the lowest cognitive level and extract the most relevant information. SCWorld seizes and synthesizes the data analyzed by Expression and presents it as a forced-directed diagram, in which topics are represented as nodes, and relationships between nodes that influence a specific topic as links. Nodes and links change in a reactive way based on what is happening in the social network in real time. We present the design of SCWorld as well as initial findings regarding its utility based on user testing of a proof-of-concept prototype.