The combination of multiple classifiers, commonly referred to as a classifier ensemble, has previously demonstrated the ability to improve classification accuracy in many application domains. As a result this area has attracted significant amount of research in recent years. The aim of this paper has therefore been to provide a state of the art review of the most well-known ensemble techniques with the main focus on bagging, boosting and stacking and to trace the recent attempts, which have been made to improve their performance. Within this paper, we present and compare an updated view on the different modifications of these techniques, which have specifically aimed to address some of the drawbacks of these methods namely the low diversity problem in bagging or the over-fitting problem in boosting. In addition, we provide a review of different ensemble selection methods based on both static and dynamic approaches. We present some new directions which have been adopted in the area of classifier ensembles from a range of recently published studies. In order to provide a deeper insight into the ensembles themselves a range of existing theoretical studies have been reviewed in the paper.

Recent discoveries in automated planning are broadening the scope of planners, from toy problems to real applications. However, applying automated planners to real-world problems is far from simple. On the one hand, the definition of accurate action models for planning is still a bottleneck. On the other hand, off-the-shelf planners fail to scale-up and to provide good solutions in many domains. In these problematic domains, planners can exploit domain-specific control knowledge to improve their performance in terms of both speed and quality of the solutions. However, manual definition of control knowledge is quite difficult. This paper reviews recent techniques in machine learning for the automatic definition of planning knowledge. It has been organized according to the target of the learning process: automatic definition of planning action models and automatic definition of planning control knowledge. In addition, the paper reviews the advances in the related field of reinforcement learning.

Ripple Down Rules (RDR) were developed in answer to the problem of maintaining medium to large rule-based knowledge systems. Traditional approaches to knowledge-based systems gave little thought to maintenance as it was expected that extensive upfront domain analysis involving a highly trained specialist, the knowledge engineer, and the time-poor domain expert would produce a complete model capturing what was in the expert’s head. The ever-changing, contextual and embrained nature of knowledge were not a part of the philosophy upon which they were based. RDR was a paradigm shift, which made knowledge acquisition and maintenance one and the same thing by incrementally acquiring knowledge as domain experts directly interacted with naturally occurring cases in their domain. Cases played an integral part of the acquisition process by motivating the capture of new knowledge, framing the context in which new knowledge would apply and ensuring that previously correctly classified cases remained so by requiring that the classification of the new case distinguish it from the system’s classification and be justified by features of the new case. RDR has moved beyond its first representation which handled single classification tasks within the domain of pathology to support multiple conclusions across a wide range of domains such as help-desk support, email classification and RoboCup and problem types including configuration, simulation, planning and natural language processing. This paper reviews the history of RDR research over the past two decades with a view to its future.

The development of a versatile, fully-capable humanoid robot as envisioned in science fiction books is one of the most challenging but interesting issues in the robotic field. Currently, existing humanoid robots are designed with different purposes and applications in mind. In humanoid robot development process, each robot is designed with various characteristics, abilities, and equipment, which influence the general structure, cost, and difficulty of development. Even though humanoid robot development is very popular, a few review papers are focusing on the design and development process of humanoid robots. Motivated by this, we present this review paper to show variations in the requirements, design, and development process and also propose a taxonomy of existing humanoid robots. It aims at demonstrating a general perspective of existing humanoid robots’ characteristics and applications. This paper includes state-of-the-art and successfully reported existing humanoid robot designs along with different robots used in various robot competitions.

This paper reviews a wide range of knowledge acquisition techniques in the context of attempts to achieve a systematic methodology. These have been poorly documented by expert system builders, who are often inclined to overvalue textbooks and the ways experts themselves claim they solve problems. No one method has a universal advantage; each has some value. Techniques should be selected to suit the domain, the task, the expert and the knowledge engineer. Knowledge acquisition involves creating a conceptual model of expert knowledge and reasoning, from analysis of data elicited by these techniques. A survey of the literature indicates increasing emphasis on tools for knowledge acquisition, used directly by experts. Several projects currently directed towards providing a proper epistemological foundation for knowledge acquisition are discussed and compared. None has yet produced a complete epistemologically sound methodology; however, recognition of the need to create a conceptual model at the knowledge level (rather than the symbol level) is an important advance. An extensive bibliography is appended.

The time evolution of aggregate economic variables, such as stock prices, is affected by market expectations of individual investors. Neoclassical economic theory assumes that individuals form expectations rationally, thus forcing prices to track economic fundamentals and leading to an efficient allocation of resources. However, laboratory experiments with human subjects have shown that individuals do not behave fully rationally but instead follow simple heuristics. In laboratory markets, prices may show persistent deviations from fundamentals similar to the large swings observed in real stock prices.

Here we show that evolutionary selection among simple forecasting heuristics can explain coordination of individual behavior, leading to three different aggregate outcomes observed in recent laboratory market-forecasting experiments: slow monotonic price convergence, oscillatory dampened price fluctuations, and persistent price oscillations. In our model, forecasting strategies are selected every period from a small population of plausible heuristics, such as adaptive expectations and trend-following rules. Individuals adapt their strategies over time, based on the relative forecasting performance of the heuristics. As a result, the evolutionary switching mechanism exhibits path dependence and matches individual forecasting behavior as well as aggregate market outcomes in the experiments. Our results are in line with recent work on agent-based models of interaction and contribute to a behavioral explanation of universal features of financial markets.

The sports domain presents a number of significant computational challenges for artificial intelligence (AI) and machine learning (ML). In this paper, we explore the techniques that have been applied to the challenges within team sports thus far. We focus on a number of different areas, namely match outcome prediction, tactical decision making, player investments, fantasy sports, and injury prediction. By assessing the work in these areas, we explore how AI is used to predict match outcomes and to help sports teams improve their strategic and tactical decision making. In particular, we describe the main directions in which research efforts have been focused to date. This highlights not only a number of strengths but also weaknesses of the models and techniques that have been employed. Finally, we discuss the research questions that exist in order to further the use of AI and ML in team sports.

Semantic technologies promise to solve many challenging problems of the present Web applications. As they achieve a feasible level of maturity, they become increasingly accepted in various business settings at enterprise level. By contrast, their usability in open environments such as the Web—with respect to issues such as scalability, dynamism and openness—still requires additional investigation. In particular, Semantic Web services have inherited the Web service communication model, which is primarily based on synchronous message exchange technology such as remote procedure call (RPC), thus being incompatible with the REST (REpresentational State Transfer) architectural model of the Web. Recent advances in the field of middleware propose ‘semantic tuplespace computing’ as an instrument for coping with this situation. Arguing that truly Web-compliant Web service communication should be based, analogously to the conventional Web, on shared access to persistently published data instead of message passing, space-based middleware introduces a coordination infrastructure by means of which services can exchange information in a time- and reference-decoupled manner. In this article, we introduce the most important approaches in this newly emerging field. Our objective is to analyze and compare the solutions proposed so far, thus giving an account of the current state-of-the-art, and identifying new directions of research and development.

The aim of this critical review paper is threefold: (a) to provide an insight on the impact of ontology engineering methodologies (OEMs) to the evolution of living and reused ontologies, (b) to update the ontology engineering (OE) community on the status and trends in OEMs and of their use in practice and (c) to propose a set of recommendations for working ontologists to consider during the life cycle of living, evolved and reused ontologies. The work outlined in this critical review paper has been motivated by the need to address critical issues on keeping ontologies alive and evolving while these are shared in wide communities. It is argued that the engineering of ontologies must follow a well-defined methodology, addressing practical aspects that would allow (sometimes wide) communities of experts and ontologists to reach consensus on developments and keep the evolution of ontologies ‘in track’. In doing so, specific collaborative and iterative tool-supported tasks and phases within a complete and evaluated ontology life cycle are necessary. This way the engineered ontologies can be considered ‘shared, commonly agreed and continuously evolved “live” conceptualizations’ of domains of discourse. Today, in the era of Linked Data and Knowledge Graphs, it is more necessary than ever not to neglect to consider the recommendations that OEMs explicitly and implicitly introduce and their implications to the evolution of living ontologies. This paper reports on the status of OEMs, identifies trends and provides recommendations based on the findings of an analysis that concerns the impact of OEMs to the status of well-known, widely used and representative ontologies.

Building ontologies in a collaborative and increasingly community-driven fashion has become a central paradigm of modern ontology engineering. This understanding of ontologies and ontology engineering processes is the result of intensive theoretical and empirical research within the Semantic Web community, supported by technology developments such as Web 2.0. Over 6 years after the publication of the first methodology for collaborative ontology engineering, it is generally acknowledged that, in order to be useful, but also economically feasible, ontologies should be developed and maintained in a community-driven manner, with the help of fully-fledged environments providing dedicated support for collaboration and user participation. Wikis, and similar communication and collaboration platforms enabling ontology stakeholders to exchange ideas and discuss modeling decisions are probably the most important technological components of such environments. In addition, process-driven methodologies assist the ontology engineering team throughout the ontology life cycle, and provide empirically grounded best practices and guidelines for optimizing ontology development results in real-world projects. The goal of this article is to analyze the state of the art in the field of collaborative ontology engineering. We will survey several of the most outstanding methodologies, methods and techniques that have emerged in the last years, and present the most popular development environments, which can be utilized to carry out, or facilitate specific activities within the methodologies. A discussion of the open issues identified concludes the survey and provides a roadmap for future research and development in this lively and promising field.

The Semantic Web anticipates integrated access to a large number of information sources on the Internet represented as Resource Description Framework (RDF). Given the large number of websites that are backed by SQL databases, methods that automate the translation of those databases to RDF are crucial. One approach, taken by a number of researchers, is to directly map the SQL schema to an equivalent Web Ontology Language (OWL) or RDF Schema representation, which in turn, implies an RDF representation for the relational data. This paper reviews this research, and derives a consolidated, overarching set of translation rules expressible as a stratified Datalog program. We present all the possible key combinations in an SQL schema and consider their implied semantic properties. We review the approaches and characterize them with respect to the scope of their coverage of SQL constructs.

Predicting or learning numeric features is called regression in the statistical literature, and it is the subject of research in both machine learning and statistics. This paper reviews the important techniques and algorithms for regression developed by both communities. Regression is important for many applications, since lots of real life problems can be modeled as regression problems. The review includes Locally Weighted Regression (LWR), rule-based regression, Projection Pursuit Regression (PPR), instance-based regression, Multivariate Adaptive Regression Splines (MARS) and recursive partitioning regression methods that induce regression trees (CART, RETIS and M5).

Limitations of sensors and the situation of a specific measurement can affect the quality of context information that is implicitly collected in pervasive environments. The lack of information about Quality of Context (QoC) can result in degraded performance of context-aware systems in pervasive environments, without knowing the actual problem. Context-aware systems can take advantage of QoC if context producers also provide QoC metrics along with context information. In this paper, we analyze QoC and present our model for processing QoC metrics. We evaluate QoC metrics considering the capabilities of sensors, circumstances of specific measurement, requirements of context consumer, and the situation of the use of context information. We also illustrate how QoC metrics can facilitate in enhancing the effectiveness and efficiency of different tasks performed by a system to provide context information in pervasive environments.

This paper reviews Machine Learning (ML), and extends and complements previous work (Kocabas, 1991; Kalkanis and Conroy, 1991). Although this paper focuses on inductive learning, it at least touches on a great many aspects of ML in general. In addition, incremental induction is also reviewed. Therefore, a general review of ML is presented, but specific detail which has been covered previously is omitted, although other relevant references are noted, and later material is commented upon.

Today's organizations require techniques for automated transformation of their large data volumes into operational knowledge. This requirement may be addressed by using event recognition systems that detect events/activities of special significance within an organization, given streams of ‘low-level’ information that is very difficult to be utilized by humans. Consider, for example, the recognition of attacks on nodes of a computer network given the Transmission Control Protocol/Internet Protocol messages, the recognition of suspicious trader behaviour given the transactions in a financial market and the recognition of whale songs given a symbolic representation of whale sounds. Various event recognition systems have been proposed in the literature. Recognition systems with a logic-based representation of event structures, in particular, have been attracting considerable attention, because, among others, they exhibit a formal, declarative semantics, they have proven to be efficient and scalable and they are supported by machine learning tools automating the construction and refinement of event structures. In this paper, we review representative approaches of logic-based event recognition and discuss open research issues of this field. We illustrate the reviewed approaches with the use of a real-world case study: event recognition for city transport management.

Argumentation is concerned with reasoning in the presence of imperfect information by constructing and weighing up arguments. It is an approach for inconsistency management in which conflict is explored rather than eradicated. This form of reasoning has proved applicable to many problems in knowledge engineering that involve uncertain, incomplete or inconsistent knowledge. This paper concentrates on different issues that can be tackled by automated argumentation systems and highlights important directions in argument-oriented research in knowledge engineering.

The Resource Description Framework (RDF) is a flexible model for representing information about resources on the Web. As a W3C (World Wide Web Consortium) Recommendation, RDF has rapidly gained popularity. With the widespread acceptance of RDF on the Web and in the enterprise, a huge amount of RDF data is being proliferated and becoming available. Efficient and scalable management of RDF data is therefore of increasing importance. RDF data management has attracted attention in the database and Semantic Web communities. Much work has been devoted to proposing different solutions to store RDF data efficiently. This paper focusses on using relational databases and NoSQL (for ‘not only SQL (Structured Query Language)’) databases to store massive RDF data. A full up-to-date overview of the current state of the art in RDF data storage is provided in the paper.

Ontology learning aims at reducing the time and efforts in the ontology development process. In recent years, several methods and tools have been proposed to speed up this process using different sources of information and different techniques. In this paper, we have reviewed 13 methods and 14 tools for semi-automatically building ontologies from texts and their relationships with the techniques each method follows. The methods have been grouped according to the main techniques followed and three groups have been identified: one based on linguistics, one on statistics, and one on machine learning. Regarding the tools, the criterion for grouping them, which has been the main aim of the tool, is to distinguish what elements of the ontology can be learned with each tool. According to this, we have identified three kinds of tools: tools for learning relations, tools for learning new concepts, and assisting tools for building up taxonomies.

Context modeling has long been acknowledged as a key aspect in a wide variety of problem domains. In this paper we focus on the combination of contextualization and personalization methods to improve the performance of personalized information retrieval. The key aspects in our proposed approach are (1) the explicit distinction between historic user context and live user context, (2) the use of ontology-driven representations of the domain of discourse, as a common, enriched representational ground for content meaning, user interests, and contextual conditions, enabling the definition of effective means to relate the three of them, and (3) the introduction of fuzzy representations as an instrument to properly handle the uncertainty and imprecision involved in the automatic interpretation of meanings, user attention, and user wishes. Based on a formal grounding at the representational level, we propose methods for the automatic extraction of persistent semantic user preferences, and live, ad-hoc user interests, which are combined in order to improve the accuracy and reliability of personalization for retrieval.

This survey introduces existing approaches to Agent Communication Languages (ACLs) and particularly Conversation Policies (CPs) which can be viewed as general constraints on the sequence of semantically coherent messages leading to a goal. Then limitations of these CPs are discussed in detail, particularly limitations on flexibility and specification. Finally, ACLs are viewed from the dialectic point of view, and some approaches are introduced in this context: some focusing on commitment-based protocols and others on dialogue-game-based protocols.