Randy Goebel

Theorist: A Logical Reasoning System for Defaults and Diagnosis

We provide an introduction to Theorist, a logic programming system that uses a uniform deductive reasoning mechanism to construct explanations of observations in terms of facts and hypotheses. Observations, facts, and possible hypotheses are each sets of logical formulas that represent, respectively, a set of observations on a partial domain, a set of facts for which the domain is a model, and a set of tentative hypotheses which may be required to provide a consistent explanation of the observations.

Local Community Identification in Social Networks

There has been much recent research on identifying global community structure in networks. However, most existing approaches require complete information of the graph in question, which is impractical for some networks, e.g. the World Wide Web (WWW). Algorithms for local community detection have been proposed but their results usually contain many outliers. In this paper, we propose a new measure of local community structure, coupled with a two-phase algorithm that extracts all possible candidates first, and then optimizes the community hierarchy. We compare our results with previous methods on real world networks such as the co-purchase network from Amazon. Experimental results verify the feasibility and effectiveness of our approach.

Discriminative Learning of Selectional Preference from Unlabeled Text

We present a discriminative method for learning selectional preferences from unlabeled text. Positive examples are taken from observed predicate-argument pairs, while negatives are constructed from unobserved combinations. We train a Support Vector Machine classifier to distinguish the positive from the negative instances. We show how to partition the examples for efficient training with 57 thousand features and 6.5 million training instances. The model outperforms other recent approaches, achieving excellent correlation with human plausibility judgments. Compared to Mutual Information, it identifies 66% more verb-object pairs in unseen text, and resolves 37% more pronouns correctly in a pronoun resolution experiment.

DBconnect: mining research community on DBLP data

Extracting information from large collections of structured, semi-structured or even unstructured data can be a considerable challenge when much of the hidden information is implicit within relationships among entities within the data. Social networks are such data collections in which relationships play a vital role in the knowledge these networks can convey. A bibliographic database is an essential tool for the research community, yet finding and making use of relationships comprised within such a social network is difficult. In this paper we introduce DBconnect, a prototype that exploits the social network coded within the DBLP database by drawing on a new random walk approach to reveal interesting knowledge about the research community and even recommend collaborations.

Extracellular Proteomes of Arabidopsis Thaliana and Brassica Napus Roots: Analysis and Comparison by MudPIT and LC-MS/MS

An important principle of the functional organization of plant cells is the targeting of proteins to specific subcellular locations. The physical location of proteins within the apoplasm/rhizosphere at the root–soil interface positions them to play a strategic role in plant response to biotic and abiotic stress. We previously demonstrated that roots of Triticum aestivum and Brassica napus exude a large suite of proteins to the apoplasm/rhizosphere [Basu et al. (1994) Plant Physiol 106:151–158; Basu et al. (1999) Physiol Plant 106:53–61]. This study is a first step to identify low abundance extracytosolic proteins from Arabidopsis thaliana and Brassica napus roots using recent advances in the field of proteomics. A total of 16 extracytosolic proteins were identified from B. napus using two-dimensional gel electrophoresis, tandem mass spectrometry (LC-MS/MS) and de novo sequencing. Another high-throughput proteomics approach, Multidimensional Protein Identification Technology (Mud PIT) was used to identify 52 extracytosolic proteins from A. thaliana. Signal peptide cleavage sites, the presence/absence of transmembrane domains and GPI modification were determined for these proteins. Functional classification grouped the extracellular proteins into different families including glycoside hydrolases, trypsin/protease inhibitors, plastocyanin-like domains, copper–zinc superoxide dismutases, gamma-thioinins, thaumatins, ubiquitins, protease inhibitor/seed storage/lipid transfer proteins, transcription factors, class III peroxidase, and plant basic secretory proteins (BSP). We have also developed an on-line, Extracytosolic Plant Proteins Database (EPPdb, http://eppdb.biology.ualberta.ca) to provide information about these extracytosolic proteins.

Visualizing and discovering web navigational patterns

Web site structures are complex to analyze. Cross-referencing the web structure with navigational behaviour adds to the complexity of the analysis. However, this convoluted analysis is necessary to discover useful patterns and understand the navigational behaviour of web site visitors, whether to improve web site structures, provide intelligent on-line tools or offer support to human decision makers. Moreover, interactive investigation of web access logs is often desired since it allows ad hoc discovery and examination of patterns not a priori known. Various visualization tools have been provided for this task but they often lack the functionality to conveniently generate new patterns. In this paper we propose a visualization tool to visualize web graphs, representations of web structure overlaid with information and pattern tiers. We also propose a web graph algebra to manipulate and combine web graphs and their layers in order to discover new patterns in an ad hoc manner.

Size-constrained tree partitioning: Approximating the multicast k-tree routing problem

In the multicast k-tree routing problem, a data copy is sent from the source node to at most k destination nodes in every transmission. The goal is to minimize the total cost of sending data to all destination nodes, which is measured as the sum of the costs of all routing trees. This problem was formulated out of optical networking and has applications in general multicasting. Several approximation algorithms, with increasing performance, have been proposed in the last several years; the most recent ones rely heavily on a tree partitioning technique. In this paper, we present a further improved approximation algorithm along the line. The algorithm has a worst-case performance ratio of 54@r+32, where @r denotes the best approximation ratio for the Steiner minimum tree problem. The proofs of the technical routing lemmas also provide some insights into why such a performance ratio could be the best possible that one can get using this tree partitioning technique.

Selecting dissimilar genes for multi-class classification, an application in cancer subtyping

BackgroundGene expression microarray is a powerful technology for genetic profiling diseases and their associated treatments. Such a process involves a key step of biomarker identification, which are expected to be closely related to the disease. A most important task of these identified genes is that they can be used to construct a classifier which can effectively diagnose disease and even recognize the disease subtypes. Binary classification, for example, diseased or healthy, in microarray data analysis has been successful, while multi-class classification, such as cancer subtyping, remains challenging.ResultsWe target on the challenging multi-class classification in microarray data analysis, especially on the cancer subtyping using gene expression microarray. We present a novel class discrimination strength vector to represent individual genes and introduce a new measurement to quantify the class discrimination strength difference between two genes. Such a new distance measure is employed in gene clustering, and subsequently the gene cluster information is exploited to select a set of genes which can be used to construct a sample classifier.We tested our method on four real cancer microarray datasets each contains multiple subtypes of cancer patients. The experimental results show that the constructed classifiers all achieved a higher classification accuracy than the previously best classification results obtained on these four datasets. Additional tests show that the selected genes by our method are less correlated and they all contribute statistically significantly to the more accurate cancer subtyping.ConclusionThe proposed novel class discrimination strength vector is a better representation than the gene expression vector, in the sense that it can be used to effectively eliminate highly correlated but redundant genes for classifier construction. Such a method can build a classifier to achieve a higher classification accuracy, which is demonstrated via cancer subtyping.

Detecting Communities in Large Networks by Iterative Local Expansion

Much structured data of scientific interest can be represented as networks, where sets of nodes or vertices are joined together in pairs by links or edges. Although these networks may belong to different research areas, there is one property that many of them do have in common: the network community structure, which means that there exists densely connected groups of vertices, with only sparser connections between groups. Identifying community structure in networks has attracted much research attention. However, most existing approaches require structure information of the graph in question to be completely accessible, which is impractical for some large networks, e.g., the World Wide Web (WWW). In this paper, we propose a community discovery algorithm for large networks that iteratively finds communities based on local information only. We compare our algorithm with previous global approaches to show its scalability. Experimental results on real world networks, such as the co-purchase network from Amazon, verify the feasibility and effectiveness of our approach.

A Sketch of Analogy as Reasoning with Equality Hypotheses

We specify a form of analogical reasoning in terms of a system of hypothetical reasoning based on mathematical logic. Our primary motivation is a deeper understanding of analogical reasoning as well as its relationship to existing logical models of nonmonotonic reasoning.