http://www.computer.org/intelligent IEEE Intelligent Systems, a bimonthly publication of the IEEE Computer Society, covers new tools, techniques, concepts, and current research and development activities in intelligent systems. The magazine serves software engineers, systems designers, information managers, knowledge engineers, and professionals in finance, manufacturing, medicine, law, and geophysical sciences. en-us Wed, 4 Jan 2012 11:00:01 GMT http://csdl.computer.org/common/images/logos/intelligent.gif List of recently published journal articles http://www.computer.org/intelligent http://doi.ieeecomputersociety.org/10.1109/MIS.2011.37 This paper presents a strategy for enhancing systems which perform tasks requiring human-level intelligence by providing them with the linguistic and cultural knowledge typically prerogative of human beings. The idea is to define a knowledge infusion process which analyzes unstructured information stored in open knowledge sources on the Web to create a memory of linguistic competencies and world facts that can be effectively exploited by the system for a deeper understanding of the information it deals with. We present OTTHO – On the Tip of my THOught – a system which implements that process for solving a challenging language game, called Guillotine, which demands knowledge covering a broad range of topics. Experiments show promising results, and our feeling is that the approach has a great potential for other more practical applications besides language games. http://doi.ieeecomputersociety.org/10.1109/MIS.2011.37 http://doi.ieeecomputersociety.org/10.1109/MIS.2010.131 Despite the increasing popularity of social information systems, there is still a general lack of research that studies the cognitive effectiveness afforded by these systems in facilitating social learning activities such as knowledge exploration, sharing, and exchange. A formal theory of the underlying individual cognitive processes of social learning will provide useful predictions on how changes in interface representations and interaction methods may impact effectiveness of these systems in facilitating knowledge exploration by multiple users. A social learning model was developed and tested against longitudinal data as participants performed knowledge exploration tasks across a period of eight weeks. Results showed that the quality of social tags and distributions of information contents directly impacted social learning. The results and the model have important implications on how Web 2.0 technologies should be designed to optimize the match between human and machine learning processes to facilitate social learning and knowledge exploration through better human-system integration. The model also demonstrates how computational models that keep track of cognitive changes of individuals can complement data-mining techniques to predict how different information cues will be interpreted and utilized by human users. http://doi.ieeecomputersociety.org/10.1109/MIS.2010.131 http://doi.ieeecomputersociety.org/10.1109/MIS.2010.108 A large portion of Web content contains geo-referenced entities. Searching for such spatial information in support of user tasks poses many technical challenges. This paper introduces a novel personalized spatial information search approach based on the Bi-directional Neural Associative Memory (BNAM) model. In this approach, search results are sensitive to user preferences. User preferences are elicited through interactive selection of spatial entities. The spatial distribution of entities of interest is based on contextual proximity, which in turn is computed with help from a taxonomy relevant to a given application domain. Our work is illustrated though a Web-based prototype covering geo-referenced entities in the city of Kyoto. http://doi.ieeecomputersociety.org/10.1109/MIS.2010.108 http://doi.ieeecomputersociety.org/10.1109/MIS.2012.1 There is a common belief that making systems more autonomous will improve the system and is therefore a desirable goal. Though small scale simple tasks can often benefit from automation, this does not necessarily generalize to more complex joint activity. When designing today’s more sophisticated systems to work closely with humans, it is important not only to consider the machine’s ability to work independently through autonomy, but also its ability to support interdependence with those involved in the joint activity. We posit that to truly improve systems and have them reach their full potential, designing systems that support interdependent activity between participants is the key. Our claim is that increasing autonomy, even in a simple and benign environment, does not always result in an improved system. We will show results from an experiment in which we demonstrate this phenomena and explain why increasing autonomy can sometimes negatively impact performance. http://doi.ieeecomputersociety.org/10.1109/MIS.2012.1 http://doi.ieeecomputersociety.org/10.1109/MIS.2011.108 We propose a supervised approach to Word Sense Disambiguation based on Neural Networks combined with Evolutionary Algorithms. An established method to automatically design the structure and learn the connection weights of Neural Networks by means of an Evolutionary Algorithm is used to evolve a neural-network disambiguator for each polysemous word, against a dataset extracted from an annotated corpus. Two distributed encoding schemes, based on the orthography of words and characterized by different degrees of information compression, have been used to represent the context in which a word occurs. The performance of such encoding schemes has been compared. The viability of the approach has been demonstrated through experiments carried out on a representative set of polysemous words. Comparison with the best entry of the Semeval-2007 competition has shown that the proposed approach is almost competitive with state-of-the-art WSD approaches. http://doi.ieeecomputersociety.org/10.1109/MIS.2011.108 http://opac.ieeecomputersociety.org/opac?year=2011&volume=26&issue=06&acronym=intelligent IEEE Intelligent Systems http://www.computer.org/portal/site/intelligent/ http://doi.ieeecomputersociety.org/10.1109/MIS.2011.93 This paper studies how people reveal private information in strategic settings in which participants need to negotiate over resources, but are uncertain about each other's objectives. The study compared two negotiation protocols which differed in whether they allowed participants to disclose their objectives in a repeated negotiation setting of incomplete information. Results show that most people agree to reveal their goals when asked, and this leads participants to more beneficial agreements. Machine learning was used to model the likelihood that people reveal their goals in negotiation, and this model was used to make goal request decisions in the game. In simulation, use of this model is shown to outperform people making the same type of decisions. These results demonstrate the benefit of this approach towards designing agents to negotiate with people under incomplete information. http://doi.ieeecomputersociety.org/10.1109/MIS.2011.93 http://doi.ieeecomputersociety.org/10.1109/MIS.2011.41 Intelligent System (IS) techniques have been continuously attracting attentions for a variety of electric power engineering applications. A recent article has summarized, from a selection of the state-of-the-art research works, the key areas in which power systems and energy markets can benefit most from IS techniques. Along with the notable capability in solving problems like electricity market simulation, market risk management, power grid planning and voltage control, the potential of IS in facilitating power system real-time stability assessment for blackout prevention has also been clearly shown in recent years. This article systematically describes the incentive, benefit, and process of constructing IS for power system real-time stability assessment, and discusses some most critical issues in the development and implementation stages. Also, on the basis of our own research output, some possible solutions to the problems are provided. Hopefully, more attention and research efforts can be drawn, to this promising while challenging topic. http://doi.ieeecomputersociety.org/10.1109/MIS.2011.41 http://doi.ieeecomputersociety.org/10.1109/MIS.2011.39 The optimal placement of actuators and sensors (OPAS) is very important in the design of intelligent structures since the placement will directly determine the control systems properties such as stability, controllability, observability etc. For such complex large space intelligent truss structures (SITS), the OPAS is a class of discrete problems that hard to solve using conventional methods. Therefore, there is an urgent need to develop a new optimization computing method. In this paper, a novel application of genetic algorithms (GA) to optimal actuators and sensors placement for SITS is presented. Optimization criteria independent of the control method is proposed and the best position of actuators and sensors is obtained. In order to verify the reliability of the optimization method, the effectiveness of fuzzy vibration control of SITS with OPAS is simulated. Numerical results show that GA is reliable, valid and superior for solving the placement optimization problem of complex structures. http://doi.ieeecomputersociety.org/10.1109/MIS.2011.39 http://doi.ieeecomputersociety.org/10.1109/MIS.2011.38 In opinion mining of product reviews, one often wants to produce a summary of opinions based on product features/attributes. However, for the same feature, people can express it with different words and phrases. To produce a meaningful summary, these words and phrases, which are domain synonyms, need to be grouped under the same feature group. This paper proposes a constrained semi-supervised learning method to solve the problem. Experimental results using reviews from five different domains show promising results. It outperforms the state-of-the-art existing methods by a large margin. http://doi.ieeecomputersociety.org/10.1109/MIS.2011.38 http://doi.ieeecomputersociety.org/10.1109/MIS.2011.36 The aim of educational systems is to assemble learning objects on a set of topics tailored to the goals and individual students' styles. Given the amount of available learning objects, the challenge of e-learning is to select the proper objects, define their relationships, and adapt their sequencing (i.e. course composition) to the specific needs, objectives and background of the student. This paper describes the general requirements for this course adaptation, the full potential of applying planning techniques on the construction of personalized e-learning routes, and how to accommodate the temporal and resource constraints to make the course applicable in a real scenario. http://doi.ieeecomputersociety.org/10.1109/MIS.2011.36 http://doi.ieeecomputersociety.org/10.1109/MIS.2011.35 Demand response (DR) is not a new concept but it is gaining a growing focus of attention in nowadays electric power systems operation and planning, with several advantages for the reliable power system functioning and for electricity prices. In this paper, price-based DR is applied to electricity consumers through the management of electricity prices. This management is based on demand elasticity and consumers are expected to react enabling to accomplish the required load reduction. The methodology is implemented in a developed DR simulator – DemSi - that uses PSCAD® for technical validation of solutions and Particle Swarm Optimization (PSO) for solution optimization. The performance of PSO is evaluated in terms of running time and obtained solutions in comparison with the Non-Linear Programming (NLP) solutions obtained in GAMS™. Case studies involving 32 and 320 consumers are used to illustrate the proposed methodology and to discuss its performance. http://doi.ieeecomputersociety.org/10.1109/MIS.2011.35 http://doi.ieeecomputersociety.org/10.1109/MIS.2011.1 In the last decade, researchers in the Online Engineering field have attempted to provide hands-on, web-based approaches for Distance Learning. The primary goal of this research is to produce Online Laboratories that serve as the eduational substitute for in situ Laboratories for Distance Learning. A limitation of existing Online Laboratories, however, is that they generally only allow a single user to be connected at a time. Since group learning activities, such as peer assistance, peer emulation, and collaborative experimental setup, are core dimensions of the traditional laboratory experience, this shortcoming is a significant pedagogical bottleneck. Recent research has focused on creating Collaborative Online Laboratories (COL) which attempt to address this shortcoming by focusing on the group awareness aspect of the laboratory learning experience. This paper discusses how group awareness can serve as a key component in replicating the collaborative aspect of learning in local laboratories. We discuss strategies for describing group awareness and how these strategies are associated both with a tutor's pedagogical objectives and in the management of the group of collaborating students. We describe an experimental system that we have developed that uses Semantic Web technologies to define a knowledge-driven system that allows researchers to describe and execute a variety of collaborative strategies for online laboratories. http://doi.ieeecomputersociety.org/10.1109/MIS.2011.1 http://doi.ieeecomputersociety.org/10.1109/MIS.2010.135 Protein kinases have played a central role in regulating many cellular pathways. A deep study into the degree of activation and inhibition of catalytic and regulatory subunits of protein kinases assists in understanding their profound effect on a cell. The inhibitors of kinase activity are a frequent cause of diseases, where kinases participate in many aspects that control cell growth, movement and death. Thus, it is critical to discover the inhibition pathways for protein kinases as well as positive patterns. This article develops an innovative methodology for negative rule association, X →¬Y for investigating the potential inhibitive regulatory correlation between the subunit isoforms of AMP-activated protein kinase (AMPK), and the stimulus factors. The rules present the pathways that have biological meaning and some were previously unknown. This not only prompts a comprehensive understanding of signalling pathways of protein kinase but also provides an attractive pharmacological target for disease treatment. http://doi.ieeecomputersociety.org/10.1109/MIS.2010.135 http://doi.ieeecomputersociety.org/10.1109/MIS.2010.134 Word sense disambiguation is the process of determining which sense of a word is used in a given context. Due to its importance in understanding semantics and many real-world applications, word sense disambiguation has been extensively studied in Natural Language Processing and Computational Linguistics. However, existing methods either narrowly focus on a few specific words due to their reliance on expensive manually annotated training text, or give only mediocre performance in real-world settings. Broad coverage and disambiguation quality are critical for real-world natural language processing applications. In this paper we present a fully automatic disambiguation method that utilizes two readily available knowledge sources: a dictionary and knowledge extracted from unannotated text. Such an automatic approach overcomes the knowledge acquisition bottleneck suffered and makes broad-coverage word sense disambiguation feasible in practice. Evaluated with two large scale WSD evaluation corpora, our system significantly outperforms the best unsupervised system and achieves the similar performance as the top-performing supervised systems. http://doi.ieeecomputersociety.org/10.1109/MIS.2010.134 http://doi.ieeecomputersociety.org/10.1109/MIS.2010.133 In this paper, we present a decision-theoretic approach to plan inference. Based on the assumption that a rational agent will adopt a plan that maximizes its expected utility, we view plan inference as reasoning about the decision-making strategy of the observed agent. Different from the previous related work, our approach explicitly takes the observed agent’s preferences into consideration, and computes the expected utilities of plans to disambiguate competing hypotheses. We use online group data to construct the domain plan library and empirically evaluate our approach in group behavior prediction. The experimental results show the effectiveness of our approach in inferring intentions and goals of entities. http://doi.ieeecomputersociety.org/10.1109/MIS.2010.133 http://doi.ieeecomputersociety.org/10.1109/MIS.2010.130 Answer ranking has significant impacts on user experience for social question answering service. The emerging challenge would be not only due to spam, malware or fraud contained in user-generated questions and answers (Q/A), but also due to the lexical and semantic gaps between the two heterogeneous collections. Existing solutions mainly focus on natural language process (NLP) techniques, generating redundant features, or finding textural clues by machine learning. They either model Q/A as independent information source by merely adding new facts and applying inference rules, or ignore any previous social knowledge that already exists and may be helpful. We assume that answers are tied to the questions with various types of latent link, and propose an analogical reasoning-based approach which learns to measure the analogy between the new question-answer(q-a) linkages and the most relevant bodies of previous knowledge which contains the subpopulation of only positive links, so that the more analogous link to the set-based analogy the higher rank assigned to the answer. We conducted our experiments based on 29.8 million Yahoo!Answer q-a threads of mono and cross-domain phenomena. The results and analysis identified the effectiveness of our approach. http://doi.ieeecomputersociety.org/10.1109/MIS.2010.130 http://doi.ieeecomputersociety.org/10.1109/MIS.2010.114 We propose a holistic or appearance based eigenfeature regularization methodology for recognizing human activities. This regularization is based on a 3-parameter based eigenmodel derived from the variances of the within-class (activity) scatter matrix computed from the intensity information appearing in activity images. Original eigenvalues are replaced by the model eigenvalues which enables us to perform discriminant evaluation in the whole eigenspace and alleviates the problems of instability, overfitting, or poor generalization. These efforts facilitate in extracting discriminative and stable low-dimensional feature representation of the activity images. Experimental results on three benchmark databases, Weizmann, INRIA-IXMAS and KTH show the superiority of our proposed approach over other popular methods. http://doi.ieeecomputersociety.org/10.1109/MIS.2010.114 http://doi.ieeecomputersociety.org/10.1109/MIS.2010.113 Today, the Web is a social, mobile, information, and communication commodity that fits in our hands and pockets. Users are using online social network sites to stay in touch with other online users and exchange useful information including health information. The online health social networking has an increasing impact on e-patients’ decisions or actions about how they treat an illness or health condition. E-patients are changing their approaches of maintaining their health, diet, exercise, or stress management. It leads them to ask new questions to their doctors when they learn about new drugs and treatments from other e-patients online. While popular online social networking sites such as Facebook and Twitters are attracting the majority of internet users, there are a few other online health social networking sites such as MedHelp, PatientsLikeMe, Inspire, and HealthCentral. As the activities in online health social networking are increasing, a major concern is the impact of these activities to the overall health outcome in a society. To understand the impact of the online health social network, identifying the influential users is helpful to extract the essence of a community. In this work, we focus on the problem of identifying the influential users in MedHelp. http://doi.ieeecomputersociety.org/10.1109/MIS.2010.113 http://doi.ieeecomputersociety.org/10.1109/MIS.2010.112 The data mining and pervasive computing technologies found in smart homes offer unprecedented opportunities for providing context-aware services, including health monitoring and assistance to individuals experiencing difficulties living independently at home. In order to provide these services, smart environment algorithms need to recognize and track activities that people normally perform as part of their daily routines. However, activity recognition has typically involved gathering and labeling large amounts of data in each setting to learn a model for activities in that setting. We hypothesize that generalized models can be learned for common activities that span multiple environment settings and resident types. We describe our approach to learning these models and demonstrate the approach using eleven CASAS datasets collected in seven environments. http://doi.ieeecomputersociety.org/10.1109/MIS.2010.112 http://doi.ieeecomputersociety.org/10.1109/MIS.2010.109 Online social networks are increasingly popular on the Internet. As people gain online expertise, their interactions become more complex, replicating behaviours from the real world and also creating new ones. Analysing and interpreting these interactions from a user perspective require techniques that go beyond statistics and integrate people and systems. Our work focuses on these issues by borrowing from Social Sciences, specifically from Activity Theory, which identifies patterns of social structures, human behaviours and uses of artefacts in social environments. This work describes these patterns as social properties with a description for automated processing and a human meaning. The analysis looks for correspondences between these properties and design information and observations on online social networks. Where a correspondence is found, the meaning of the property is useful for interpreting observations in terms of human behaviour patterns. A case study on a recommendation service illustrates the approach. http://doi.ieeecomputersociety.org/10.1109/MIS.2010.109 http://doi.ieeecomputersociety.org/10.1109/MIS.2010.92 We present an algorithm for the automated external calibration (localization) of a network of cameras with non-overlapping fields of view, a type of network that is becoming widespread for monitoring large environments. The proposed algorithm exploits trajectory observations in each view and works iteratively on camera pairs. First outliers are identified and removed from each camera observations. Next, spatio-temporal information derived from the available trajectory is used to estimate unobserved trajectory segments in areas uncovered by the cameras. The unobserved trajectory estimates are used to estimate the relative position of each camera pair, whereas the exit-entrance direction of each object is used to estimate their relative orientation. The process continues and iteratively approximates the configuration of all cameras with respect to each other. Finally, we refine the initial configuration estimates with bundle adjustment, based on the observed and estimated trajectory segments. The accuracy and the robustness of the proposed approach are demonstrated using both simulated and real datasets, and compared with state-of-the-art approaches. http://doi.ieeecomputersociety.org/10.1109/MIS.2010.92