Machine learning techniques are gaining prevalence in the production of a wide range of classifiers for complex real-world applications with nonuniform testing and misclassification costs. The increasing complexity of these applications poses a real challenge to resource management during learning and classification. In this work we introduce ACT (anytime cost-sensitive tree learner), a novel framework for operating in such complex environments. ACT is an anytime algorithm that allows learning time to be increased in return for lower classification costs. It builds a tree top-down and exploits additional time resources to obtain better estimations for the utility of the different candidate splits. Using sampling techniques, ACT approximates the cost of the subtree under each candidate split and favors the one with a minimal cost. As a stochastic algorithm, ACT is expected to be able to escape local minima, into which greedy methods may be trapped. Experiments with a variety of datasets were conducted to compare ACT to the state-of-the-art cost-sensitive tree learners. The results show that for the majority of domains ACT produces significantly less costly trees. ACT also exhibits good anytime behavior with diminishing returns.

Analyzing 0-1 matrices is one of the main themes in data mining. Techniques such as clustering or mixture modelling, matrix decomposition techniques such as PCA, ICA, and NMR, and Bayesian all aim to give an answer to the informal question: "Where does the matrix come from?" These approaches aim at describing a probabilistic generative model that describes the observed matrix well. In this paper we consider yet another way of answering the question "Where does a 0-1 matrix M come from?" In our model, the matrix M of size n m is considered to arise from initiators, certain few entries that are initially 1. The initiators propagate their 1's by following the links of a directed influence graph G (represented by an n n adjacency matrix). We denote the initiator matrix of size n m by N and we use G (of size n n) to refer both to the directed graph between the rows of M and as well as its adjacency matrix. Then, we believe that the structure of N and G can tell how a matrix M has been created.

Many real-world applications produce networked data such as the world-wide web (hypertext documents connected via hyperlinks), social networks (for example, people connected by friendship links), communication networks (computers connected via communication links) and biological networks (for example, protein interaction networks). A recent focus in machine learning research has been to extend traditional machine learning classification techniques to classify nodes in such networks. In this article, we provide a brief introduction to this area of research and how it has progressed during the past decade. We introduce four of the most widely used inference algorithms for classifying networked data and empirically compare them on both synthetic and real-world data.

Institute, Japan) serves as IJCAI-09 March 11, 2009: IAAI Notification of acceptance or rejection Conference Chair.

The Seventeenth International Conference on Automated Planning and Scheduling (ICAPS-07) was held in Providence, Rhode Island in September 2007. It covered the latest theoretical and practical advances in planning and scheduling. The conference was co-located with the Thirteenth International Conference on Principles and Practice of Constraint Programming (CP-07). The program consisted of tutorials, workshops, system demonstrations, a doctoral consortium, and three days of technical presentations mostly in parallel sessions. ICAPS-07 also hosted the second edition of the International Competition on Knowledge Engineering for Planning and Scheduling. This report describes the conference in more detail.

The Association for the Advancement of Artificial Intelligence (AAAI) was pleased to present the AAAI 2008 Spring Symposium Series, held Wednesday through Friday, March 26–28, 2008 at Stanford University, California. The titles of the eight symposia were as follows: (1) AI Meets Business Rules and Process Management, (2) Architectures for Intelligent Theory-Based Agents, (3) Creative Intelligent Systems, (4) Emotion, Personality, and Social Behavior, (5) Semantic Scientific Knowledge Integration, (6) Social Information Processing, (7) Symbiotic Relationships between Semantic Web and Knowledge Engineering, (8) Using AI to Motivate Greater Participation in Computer Science The goal of the AI Meets Business Rules and Process Management AAAI symposium was to investigate the various approaches and standards to represent business rules, business process management and the semantic web with respect to expressiveness and reasoning capabilities. The focus of the Architectures for Intelligent Theory-Based Agents AAAI symposium was the definition of architectures for intelligent theory-based agents, comprising languages, knowledge representation methodologies, reasoning algorithms, and control loops. The Creative Intelligent Systems Symposium included five major discussion sessions and a general poster session (in which all contributing papers were presented). The purpose of this symposium was to explore the synergies between creative cognition and intelligent systems. The goal of the Emotion, Personality, and Social Behavior symposium was to examine fundamental issues in affect and personality in both biological and artificial agents, focusing on the roles of these factors in mediating social behavior. The Semantic Scientific Knowledge Symposium was interested in bringing together the semantic technologies community with the scientific information technology community in an effort to build the general semantic science information community. The Social Information Processing's goal was to investigate computational and analytic approaches that will enable users to harness the efforts of large numbers of other users to solve a variety of information processing problems, from discovering high-quality content to managing common resources. The goal of the Symbiotic Relationships between the Semantic Web and Software Engineering symposium was to explore how the lessons learned by the knowledge-engineering community over the past three decades could be applied to the bold research agenda of current workers in semantic web technologies. The purpose of the Using AI to Motivate Greater Participation in Computer Science symposium was to identify ways that topics in AI may be used to motivate greater student participation in computer science by highlighting fun, engaging, and intellectually challenging developments in AI-related curriculum at a number of educational levels. Technical reports of the symposia were published by AAAI Press.

Citizens, both young and feeds, and semistructured metadata old, are also discovering how social media in the form of extensible markup language technology can improve their lives and (XML) and resource description give them more voice in the world. We they provide more useful, trustworthy, begin by describing an overarching task of and reliable. Pursuing this task uncovers It differs, however, in ways a number of problems that must be addressed, that affect how it should be modeled, analyzed, three of which we describe in and exploited. The first is recognizing spam model for the general web is as a directed graph of web pages with undifferentiated in the form of spam blogs (splogs) and links between pages. The second is developing has a much richer network structure more effective techniques to recognize in that there are more types of nodes the social structure of blog communities. For example, the abstract model for the underlying blog people who contribute to blogs and au-network structure and how it evolves. Figure 2 shows a hypothetical blog graph and its corresponding flow of information in the influence graph. Studies on influence in social networks and collaboration graphs have typically focused on the task of identifying key individuals who play an important role in propagating information. This is similar to finding authoritative pages on the web.

A large class of systems of biological and technological relevance can be described as analog networks, that is, collections of dynamical devices interconnected by links of varying strength. Some examples of analog networks are genetic regulatory networks, metabolic networks, neural networks, analog electronic circuits, and control systems. Analog networks are typically complex systems which include nonlinear feedback loops and possess temporal dynamics at different time scales. Both the synthesis and reverse engineering of analog networks are recognized as knowledge-intensive activities, for which few systematic techniques exist. In this paper we will discuss the general relevance of the analog network concept and describe an evolutionary approach to the automatic synthesis and the reverse engineering of analog networks. The proposed approach is called analog genetic encoding (AGE) and realizes an implicit genetic encoding of analog networks. AGE permits the evolution of human-competitive solutions to real-world analog network design and identification problems. This is illustrated by some examples of application to the design of electronic circuits, control systems, learning neural architectures, and the reverse engineering of biological networks.

In many cooperative multiagent domains, the effect of local interactions between agents can be compactly represented as a network structure. Given that agents are spread across such a network, agents directly interact only with a small group of neighbors. A distributed constraint optimization problem (DCOP) is a useful framework to reason about such networks of agents. Given agents’ inability to communicate and collaborate in large groups in such networks, we focus on an approach called k-optimality for solving DCOPs. In this approach, agents form groups of one or more agents until no group of k or fewer agents can possibly improve the DCOP solution; we define this type of local optimum, and any algorithm guaranteed to reach such a local optimum, as k-optimal. The article provides an overview of three key results related to koptimality. The first set of results gives worst-case guarantees on the solution quality of k-optima in a DCOP. These guarantees can help determine an appropriate k-optimal algorithm, or possibly an appropriate constraint graph structure, for agents to use in situations where the cost of coordination between agents must be weighed against the quality of the solution reached. The second set of results gives upper bounds on the number of k-optima that can exist in a DCOP. These results are useful in domains where a DCOP must generate a set of solutions rather than a single solution. Finally, we sketch algorithms for k-optimality and provide some experimental results for 1-, 2- and 3-optimal algorithms for several types of DCOPs.

This introduction to AI Magazine's Special Issueon Networks and AI summarizes the seven articles in thespecial issue by characterizing the nature of thenetworks that are the focus of each of the papers.A short tutorial on graph theory and network structuresis included for those less familiar with the topic.