Abstract. Scientific literature with rich metadata can be represented as a labeled directed graph. This graph representation enables a number of scientific tasks such as ad hoc retrieval or named entity recognition (NER) to be formulated as typed proximity queries in the graph. One popular proximity measure is called Random Walk with Restart (RWR), and much work has been done on the supervised learning of RWR measures by associating each edge label with a parameter. In this paper, we describe a novel learnable proximity measure which instead uses one weight per edge label sequence: proximity is defined by a weighted combination of simple “path experts”, each corresponding to following a particular sequence of labeled edges. Experiments on eight tasks in two subdomains of biology show that the new learning method significantly outperforms the RWR model (both trained and untrained). We also extend the method to support two additional types of experts to model intrinsic properties of entities: query-independent experts, which generalize the PageRank measure, and popular entity experts which allow rankings to be adjusted for particular entities that are especially important.

Collaborative research has been increasingly popular and important in academic circles. However, there is no open platform available for scholars or scientists to effectively discover potential collaborators. This paper discusses Collab-Seer, an open system to recommend potential research collaborators for scholars and scientists. CollabSeer discovers collaborators based on the structure of the coauthor network and a user’s research interests. Currently, three different network structure analysis methods that use vertex similarity are supported in CollabSeer: Jaccard similarity, cosine similarity, and our relation strength similarity measure. Users can also request a recommendation by selecting a topic of interest. The topic of interest list is determined by CollabSeer’s lexical analysis module, which analyzes the key phrases of previous publications. The CollabSeer system is highly modularized making it easy to add or replace the network analysis module or users ’ topic of interest analysis module. CollabSeer integrates the results of the two modules to recommend collaborators to users. Initial experimental results over the a subset of the CiteSeerX database shows that CollabSeer can efficiently discover prospective collaborators.

Automatic recommendation of citations for a manuscript is highly valuable for scholarly activities since it can substantially improve the efficiency and quality of literature search. The prior techniques placed a considerable burden on users, who were required to provide a representative bibliography or to mark passages where citations are needed. In this paper we present a system that considerably reduces this burden: a user simply inputs a query manuscript (without a bibliography) and our system automatically finds locations where citations are needed. We show that the naïve approaches do not work well due to massive noise in the document corpus. We produce a successful approach by carefully examining the relevance between segments in a query manuscript and the representative segments extracted from a document corpus. An extensive empirical evaluation using the CiteSeerX data set shows that our approach is effective.

Action Science Explorer (ASE) is a tool designed to support users in rapidly generating readily consumable summaries of academic literature. The authors describe ASE and report on how early formative evaluations led to a mature system evaluation, consisting of an in-depth empirical evaluation with 4 domain expert participants. The user study tasks were of two types: predefined tasks to test system performance in common scenarios, and user-defined tasks to test the system’s usefulness for custom exploration goals. This paper concludes by describing ASE’s attribute ranking capability which is a novel contribution for exploring scientific literature networks. It makes design recommendations to: give the users control over which documents to explore, easyto-understand metrics for ranking documents, and overviews of the document set in coordinated views along with detailson-demand of specific papers.

Hierarchical taxonomies provide a multi-level view of large document collections, allowing users to rapidly drill down to finegrained distinctions in topics of interest. We show that automatically induced taxonomies can be made more robust by combining text with relational links. The underlying mechanism is a Bayesian generative model in which a latent hierarchical structure explains the observed data — thus, finding hierarchical groups of documents with similar word distributions and dense network connections. As a nonparametric Bayesian model, our approach does not require pre-specification of the branching factor at each non-terminal, but finds the appropriate level of detail directly from the data. Unlike many prior latent space models of network structure, the complexity of our approach does not grow quadratically in the number of documents, enabling application to networks with more than ten thousand nodes. Experimental results on hypertext and citation network corpora demonstrate the advantages of our hierarchical, multimodal approach.

Investigators frequently need to quickly learn new research domains in order to advance their research. This thesis presents five contributions to understanding how software helps researchers explore scientific literature networks. (1) A taxonomy which summarizes capabilities in existing bibliography tools, revealing patterns of capabilities by system type. (2) Six participants in two user studies evaluate Action Science Explorer (ASE), which is designed to create surveys of scientific literature and integrates visualizations and statistics. Users found document-level statistics and attribute rankings to be convenient when beginning literature exploration. (3) User studies also identify users ’ questions when exploring academic literature, which include examining the evolution of a field, identifying author relationships, and searching for review papers. (4) The evaluations suggest shortcomings of ASE, and this thesis outlines improvements to ASE and lists user requirements for bibliographic exploration. (5) I recommend strategies for evaluating bibliographic exploration tools based on experiences evaluating ASE. UNDERSTANDING SCIENTIFIC LITERATURE NETWORKS:

Abstract—Action Science Explorer (ASE) is a tool designed to support users in rapidly generating readily consumable summaries of academic literature. It uses citation network visualization, ranking and filtering papers by network statistics, and automatic clustering and summarization techniques. We describe how early formative evaluations of ASE led to a mature system evaluation, consisting of an in-depth empirical evaluation with four domain experts. The evaluation tasks were of two types: predefined tasks to test system performance in common scenarios, and user-defined tasks to test the system’s usefulness for custom exploration goals. The primary contribution of this paper is a validation of the ASE design and recommendations to provide: easy-to-understand metrics for ranking and filtering documents, user control over which document sets to explore, and overviews of the document set in coordinated views along with detailson-demand of specific papers. We contribute a taxonomy of features for literature search and exploration tools and describe exploration goals identified by our participants. Keywords-Empirical evaluation; graphical user interfaces; information visualization; literature exploration I.