Barbara Rosario

Multi-way Relation Classification: Application to Protein-Protein Interactions

We address the problem of multi-way relation classification, applied to identification of the interactions between proteins in bioscience text. A major impediment to such work is the acquisition of appropriately labeled training data; for our experiments we have identified a database that serves as a proxy for training data. We use two graphical models and a neural net for the classification of the interactions, achieving an accuracy of 64% for a 10-way distinction between relation types. We also provide evidence that the exploitation of the sentences surrounding a citation to a paper can yield higher accuracy than other sentences.

The Descent of Hierarchy, and Selection in Relational Semantics

In many types of technical texts, meaning is embedded in noun compounds. A language understanding program needs to be able to interpret these in order to ascertain sentence meaning. We explore the possibility of using an existing lexical hierarchy for the purpose of placing words from a noun compound into categories, and then using this category membership to determine the relation that holds between the nouns. In this paper we present the results of an analysis of this method on two-word noun compounds from the biomedical domain, obtaining classification accuracy of approximately 90%. Since lexical hierarchies are not necessarily ideally suited for this task, we also pose the question: how far down the hierarchy must the algorithm descend before all the terms within the subhierarchy behave uniformly with respect to the semantic relation in question? We find that the topmost levels of the hierarchy yield an accurate classification, thus providing an economic way of assigning relations to noun compounds.

A synthetic agent system for Bayesian modeling of human interactions

When building statistical machine learning models from real data one of the most frequently encountered di culties is the limited amount of training data compared to what is needed by the speci c learning architecture In order to deal with this problem we have developed a synthetic simulated agent training system that let us develop exible prior models for recognizing human interactions in a pedestrian visual surveillance task We demonstrate the ability to use these prior models to accurately classify real human behaviors and interactions with no additional tuning or training