Barry Schiffman

Inferring temporal ordering of events in news

This paper describes a domain-independent, machine-learning based approach to temporally anchoring and ordering events in news. The approach achieves 84.6% accuracy in temporally anchoring events and 75.4% accuracy in partially ordering them.

Integration of Visual and Text-Based Approaches for the Content Labeling and Classification of Photographs

Annotating photographs automatically with content descriptions facilitates organization, storage, and search over visual information. We present an integrated approach for scene classi cation that combines image-based and text-based approaches. On the text side, we use the text accompanying an image in a novel TF*IDF vector-based approach to classi cation. On the image side, we present a novel OF*IIF (object frequency) vector-based approach to classi cation. Objects are de ned by clustering of segmented regions of training images. The image based OF*IIF approach is synergistic with the text based TF*IDF approach. By integrating the TF*IDF approach and the OF*IIF approach, we achieved a classi cation accuracy of 86%. This is an improvement of approximately 12% over existing image classi ers, an improvement of approximately 3% over the TF*IDF image classi er based on textual information, and an improvement of approximately 4% over the OF*IIF image classi er based on visual information.

Producing Biographical Summaries: Combining Linguistic Knowledge with Corpus Statistics

We describe a biographical multi-document summarizer that summarizes information about people described in the news. The summarizer uses corpus statistics along with linguistic knowledge to select and merge descriptions of people from a document collection, removing redundant descriptions. The summarization components have been extensively evaluated for coherence, accuracy, and non-redundancy of the descriptions produced.

Columbia University at DUC 2004

We describe our participation in tasks 2, 4 and 5 of the DUC 2004 evaluation. For each task, we present the system(s) used, focusing on novel and newly developed aspects. We also analyze the results of the human and automatic evaluations.

Question Answering Using Integrated Information Retrieval and Information Extraction

This paper addresses the task of providing extended responses to questions regarding specialized topics. This task is an amalgam of information retrieval, topical summarization, and Information Extraction (IE). We present an approach which draws on methods from each of these areas, and compare the effectiveness of this approach with a query-focused summarization approach. The two systems are evaluated in the context of the prosecution queries like those in the DARPA GALE distillation evaluation.

Columbia's newsblaster: new features and future directions

Columbias Newsblaster tracking and summarization system is a robust system that clusters news into events, categorizes events into broad topics and summarizes multiple articles on each event. Here we outline our most current work on tracking events over days, producing summaries that update a user on new information about an event, outlining the perspectives of news coming from different countries and clustering and summarizing non-English sources.

Context and Learning in Novelty Detection

We demonstrate the value of using context in a new-information detection system that achieved the highest precision scores at the Text Retrieval Conferences Novelty Track in 2004. In order to determine whether information within a sentence has been seen in material read previously, our system integrates information about the context of the sentence with novel words and named entities within the sentence, and uses a specialized learning algorithm to tune the system parameters.

Building a resource for evaluating the importance of sentences

This paper will introduce a new lexical resource for measuring the importance of short segments of text, such as sentences. The resource, a list of words compiled automatically from a large background corpus of news articles, can provide evidence that a text segment is globally important, that is intrinsically interesting, not only interesting in relation to a specified topic or set of documents.

Columbia at the Document Understanding Conference 2003

The Columbia Summarizer for DUC 2003, Task 2, isbased on the multi-document summarization system thatwe developed for DUC 2002 (McKeown et al., 2002). Ituses different summarization strategies depending on thetype of documents in the input set. Four different strate-gies are used, one for single events, one for multiple re-lated events, one for biographies and one for discussionof an issue with related events. The summarization strat-egy encoded in M

Experiments in automated lexicon building for text searching

This paper describes experiments in the automatic construction of lexicons that would be useful in searching large document collections for text fragments that address a specific information need, such as an answer to a question.