When analysing social media conversations, in search of the public opinion about an unfolding event that is be- ing discussed in real-time (e.g., presidential debates, major speeches, etc.), it is important to distinguish between two groups of participants: opinion-makers and opinion-holders. To address this problem, we propose a supervised machine-learning approach, which uses inexpensively acquired labeled data from monothematic Twitter accounts to learn a binary classifier for the labels “political account” (opinion-makers) and “non-political account” (opinion-holders). While the classifier has a 83% accuracy on individual tweets, when applied to the last 200 tweets from accounts of a set of 1000 Twitter users, it classifies accounts with a 97% accuracy. This high accuracy derives from our decision to incorporate information about classifier probability into the classification. Our work demonstrates that machine learning algorithms can play a critical role in improving the quality of social media analytics and understanding, whose importance is increasing as social media adoption becomes widespread.

Protecting our nation’s ports is a critical challenge for homeland security and requires the research, development and deployment of new technologies that will allow for the efficient securing of shipments entering this country. Most approaches look only at statistical irregularities in the attributes of the cargo, and not at the relationships of this cargo to others. However, anomalies detected in these relationships could add to the suspicion of the cargo, and therefore improve the accuracy with which we detect suspicious cargo. This paper proposes an improvement in our ability to detect suspicious cargo bound for the U.S. through a graph-based anomaly detection approach. Using anonymized data received from the Department of Homeland Security, we demonstrate the effectiveness of our approach and its usefulness to a homeland security analyst who is tasked with uncovering illegal and potentially dangerous cargo shipments.

Maritime threat detection is a challenging problem because maritime environments can involve a complex combination of concurrent vessel activities, and only a small fraction of these may be irregular, suspicious, or threatening. Previous work on this task has been limited to analyses of single vessels using simple rule-based models that alert watchstanders when a proximity threshold is breached. We claim that Probabilistic Graphical Models (PGMs) can be used to more effectively model complex maritime situations. In this paper, we study the performance of PGMs for detecting (small boat) maritime attacks. We describe three types of PGMs that vary in their representational expressiveness and evaluate them on a threat recognition task using track data obtained from force protection naval exercises involving unmanned sea surface vehicles. We found that the best-performing PGMs can outperform the deployed rule-based approach on these tasks, though some PGMs require substantial engineering and are computationally expensive.

We study the computational complexity of optimal bribery and manipulation schemes for sports tournaments with uncertain information: cup; challenge or caterpillar; and round robin. Our results carry over to the equivalent voting rules: sequential pair-wise elections, cup, and Copeland, when the set of candidates is exactly the set of voters. This restriction creates new difficulties for most existing algorithms. The complexity of bribery and manipulation are well studied, almost always assuming deterministic information about votes and results. We assume that for candidates i and j the probability that i beats j and the costs of lowering each probability by fixed increments are known to the manipulators. We provide complexity analyses for cup, challenge, and round robin competitions ranging from polynomial time to NP^PP. This shows that the introduction of uncertainty into the reasoning process drastically increases the complexity of bribery problems in some instances.

In this paper, we investigate the application of data mining to existing techniques for quality control/anomaly detection on weather sensor observations. Specifically we adapt the popular Barnes Spatial interpolation method to use time-series distance rather than spatial distance to develop an online algorithm that uses readings from similar stations based on current and historical observations for interpolation and we demonstrate that this new algorithm exhibits less model error than the Barnes Spatial interpolation-based method. We focus on interpolation, which is a basis for this popular quality control method and other related methods, and examine a dataset of over 233 million temperature observations from California and surrounding areas. Our approach shows improved performance as indicated by mean squared error reduced by approximately one half for predicted values versus reported values.

Bill Swartout Introduced by Alan Kay at XEROX PARC in the 1970's, the desktop metaphor, which was later adopted in the Macintosh and Windows operating systems, has become the primary way we think about interacting with computers. Over the last decade, we have been developing sophisticated virtual humans at the USC Institute for Creative Technologies.

Recommender systems assist users in the daunting task of sifting through large amounts of data in order to select relevant information or items. Common examples include consumer products and services, such as for songs, books, articles, etc. Unfortunately, such systems may be subject to attack by malicious users who want to manipulate the system’s recommendations to suit their needs: to promote their own (or demote a competitor’s) product/service, or to cause disruption in the recommender system. Attacks can cause the recommender system to become unreliable and untrustworthy, resulting in user dissatisfaction. Developers already face tradeoffs in system efficiency and accuracy, and designing for robustness adds an additional dimension for consideration. In this paper, we show how the underlying implementation choices for item-based and user-based Collaborative Filtering recommender systems can affect the accuracy and robustness of recommender systems. We also show how accuracy and robustness can change over a system’s lifetime by analyzing a set of temporal snapshots from system usage over time. Results provide insight into some of the tradeoffs between robustness and accuracy that operators may need to consider in development and evaluation.

We consider a cross-cultural interaction scenario where a group of soldiers assisted by robots interact with local vendors in a market place. We develop a model to quantify, analyze and predict the perception of the actions of the soldiers and the robot by the local population. The model assumes that humans are considering collections of concrete and intangible values which are not, in general, directly and linearly convertible into each other. We argue that satisfactory modeling accuracy can be achieved by restricting the considered intangibles to a small set of {\em culture sanctioned social values}. For these values, the culture provides a name, calculation methods, as well as associated rules of conduct. We validate our model by comparing the predicted values with the judgment of a large group of human observers cognizant of the modeled culture. We use the model to evaluate the tradeoffs between several long term strategies to maintain security as well as to increase the trust and goodwill of the local population.

Different studies provide evidence that the computational psycholinguistic algorithm called Latent Semantic Analysis (LSA) allows measuring local and global coherence in texts similarly to human evaluation (Foltz, Kintsch, Landauer 1998; McNamara, Cai & Louwerse 2007; McCarthy, Briner, Rus, & McNamara, 2007; McNamara, Louwerse & Jeuniaux 2009; Louwerse, McCarthy & Graesser 2010). The texts used in all these studies are written in English and correspond to scientific and literary texts. In Spanish, there are some studies using LSA that measure the semantic similarity between texts in automatic summary assessment (Pérez, Alfonseca, Rodríguez, Gliozzo, Strapparava & Magnini 2005; León, Olmos, Escudero, Cañas & Salmerón 2006; Venegas 2007, 2009, 2011); however, automatic measurement of coherence in Spanish has not yet been sufficiently investigated. The present study aimed at identifying a global and local coherence profile in a corpus of speeches in Spanish of three Latin American Heads-of-States (Perón, Castro and Pinochet), using Latent Semantic Analysis. Local coherence is calculated through the measurement of implicit semantic similarity between adjacent sentences and global coherence through the measurement of the similarity among the semantic content of the paragraphs. The corpus under analysis corresponds to a sample of 107 speeches. The semantic space was built using a multi-register corpus and it is available through the “Interface for the measurement of lexical-semantic similarity” in the El Grial interface (www.elgrial.cl). Results showed a systematic difference between the speeches of the Heads-of-State in terms of both local and global coherence. The Bonferroni analysis established an effect that distinguishes Perón’s speeches from Pinochet’s and Castro’s speeches. This results show that Perón’s speeches are more topically related than the other leaders’, probably due to a discourse strategy to persuade voters. The identification of a profile of coherence might be relevant to predict cues of government discourse styles.

Chairs and committees of the FLAIRS-25 conference.