Collaboration between research scientists, particularly those with diverse backgrounds, is a driver of scientific innovation. However, finding the right collaborator is often an unscientific process that is subject to chance. This paper explores recommending collaborators based on repeating patterns of previous successful collaboration experiences, what we term prototypical collaborations. We investigate a method for discovering such prototypes to use them as a basis to guide the recommendation of new collaborations. To this end, we also examine two methods for matching collaboration seekers to these prototypical collaborations. Our initial studies reveal that though promising, improving collaborations through recommendation is a complex goal.

As the amount of information available to researchers grows at an increasing rate, it becomes much more difficult to find relevant resources. An approach taken by several authoritative bodies, such as the Association for Computing Machinery and the U.S. National Library of Medicine, is the introduction of a classification scheme. However, even the most modern schemes are not capable of adequately distinguishing one research paper from another, due mainly to their broad generality. This paper describes a methodology for building a much narrower, specialized classification scheme focused on the area of Cased-Based Reasoning in the Health Sciences. It is derived from thorough analysis of the field, but with a framework that can be adapted to other areas. Using a tiered approach to further subdivide systems into more specific classes according to criteria specific to this particular field, this classification scheme affords interdisciplinary search, which is generally left out of generic indexing systems. This paper presents the resulting classification scheme and showcases its usefulness for classifying and tracking the evolution of research.

Generalized cases are cases that cover a subspace rather than a point in the problem-solution space. Attribute dependent generalized cases are a subclass of generalized cases, which cause a high computational complexity during similarity assessment. We present a new approach for an efficient index-based retrieval of such generalized cases by an improved kd-tree approach. The experimental evaluation demonstrates a significant improvement in retrieval efficiency compared to previous methods.

This paper discusses the state of the art in CBR ontologies from the perspective of one developing an improved system for case-based legal reasoning. The paper proposes three specific roles for a CBR ontology and illustrates them in the context of the intended output of the new system: a legal classroom discussion of how to decide a case featuring hypothetical reasoning and abstract analogies. The paper distills the ontological requirements for modeling the example’s case-based arguments and assesses whether current research can meet those requirements. The concrete example helps to focus on and define goals for improving CBR ontologies.

Following successful special tracks on case-based reasoning at FLAIRS over the past seven years, we invited papers for the Eighth Special Track on CBR at the 22nd International FLAIRS Conference. Case-based reasoning is an AI problem solving and analysis methodology that retrieves and adapts previous experiences to fit new contexts. This forum is intended to gather AI researchers and practitioners with an interest in CBR to present and discuss developments in CBR theory and application. Submission topics included foundations of CBR; methods for CBR (such as representation, indexing, retrieval, adaptation); evaluation methods for CBR systems and integrations; practical applications of CBR; textual CBR; CBR and creativity; CBR and design; distributed CBR; case based maintenance; spatiotemporal CBR; CBR in the health sciences; CBR integrations; case based planning; and CBR and games. The invited speaker for the special track for 2009 is Ashok Goel from the Georgia Institute of Technology, USA.

Rare category detection is an open challenge for active learning, especially in the de-novo case (no labeled examples), but of significant practical importance for data mining - e.g.

Rare category detection is an open challenge for active learning, especially in the de-novo case (no labeled examples), but of significant practical importance for data mining - e.g.

Rare category detection is an open challenge for active learning, especially in the de-novo case (no labeled examples), but of significant practical importance for data mining - e.g.

Can we leverage learning techniques to build a fast nearest-neighbor (NN) retrieval data structure? We present a general learning framework for the NN problem in which sample queries are used to learn the parameters of a data structure that minimize the retrieval time and/or the miss rate. We explore the potential of this novel framework through two popular NN data structures: KD-trees and the rectilinear structures employed by locality sensitive hashing. We derive a generalization theory for these data structure classes and present simple learning algorithms for both. Experimental results reveal that learning often improves on the already strong performance of these data structures.

The $k$th-nearest neighbor rule is arguably the simplest and most intuitively appealing nonparametric classification procedure. However, application of this method is inhibited by lack of knowledge about its properties, in particular, about the manner in which it is influenced by the value of $k$; and by the absence of techniques for empirical choice of $k$. In the present paper we detail the way in which the value of $k$ determines the misclassification error. We consider two models, Poisson and Binomial, for the training samples. Under the first model, data are recorded in a Poisson stream and are "assigned" to one or other of the two populations in accordance with the prior probabilities. In particular, the total number of data in both training samples is a Poisson-distributed random variable. Under the Binomial model, however, the total number of data in the training samples is fixed, although again each data value is assigned in a random way. Although the values of risk and regret associated with the Poisson and Binomial models are different, they are asymptotically equivalent to first order, and also to the risks associated with kernel-based classifiers that are tailored to the case of two derivatives. These properties motivate new methods for choosing the value of $k$.