About
Comparing and contrasting is an important strategy people employ to understand new situations and create solutions for new problems. Similar events can provide hints for problem solving, as well as larger contexts for understanding the specific circumstances of an event. Lessons can be learned from past experience, insights can be gained about the new situation from familiar examples, and trends can be discovered among similar events. As the largest knowledge base for human beings, the Web provides both an opportunity and a challenge to discover comparable cases in order to facilitate situation analysis and problem solving. In this paper, we present Compare&Contrast, a system that uses the Web to discover comparable cases for news stories, documents about similar situations but involving distinct entities. The system analyzes a news story given by the user and builds a model of the story. With the story model, the system dynamically discovers entities comparable to the main entity in the original story and uses these comparable entities as seeds to retrieve web pages about comparable cases. The system is domain independent, does not require any domain-specific knowledge engineering efforts, and deals with the complexity of unstructured text and noise on the web in a robust way. We evaluated the system with an experiment on a collection of news articles and a user study.
Citation
Liu, J., Birnbaum, L. and Wagner E. 2007. Compare&Contrast: Using the Web to Discover Comparable Cases for News Stories. In Proceedings of the 16th International Conference on World Wide Web (Banff, Alberta, Canada, May 8 - 12, 2007). WWW '07. ACM Press, New York, NY.
Related Papers
MakeMyPage: Social Media Meets Automatic Content Generation
Rich Interfaces for Reading News on the Web
Categorizing Blogger’s Interests Based on Short Snippets of Blog Posts
Pivot: Automatically Offering Information and Services to Real-World Shoppers
LocalSavvy: Aggregating Local Points of View about News Issues
What Do They Think? Aggregating Local Views about News Events and Topics
Compare&Contrast: Using the Web to Discover Comparable Cases for News Stories
Collaborative Reasoning and Collaborative Ontology Development in CRAFT
Between Ontology and Folksonomy: A Study of Collaborative and Implicit Ontology Evolution
Reasoning Through Search: A Novel Approach to Sentiment Classification
Measuring Semantic Similarity between Named Entities by Searching the Web Directory
Learning to Gesture: Applying animations To Spoken Text
TagAssist: Automatic Tag Suggestion for Blog Posts
Context Transformations for Just-in-time Retrieval: Adapting the Watson System to User Needs
Buzz: Telling Compelling Stories
Using Explicit Semantic Models to Track Situations across News Articles
Creating Polite Agents: 5 Heuristics for User Experience Design
Computational Support for Compelling Story Telling
Believable Performance Agents for Interactive Conversations
Domain Specific Affective Classification of Documents
Network Arts: Defining Emotional Interaction in Media Arts
MusicStory: a Personalized Music Video Creator
Analogy, Intelligent IR, and Knowledge Integration for Intelligence Analysis
Affective Behaviors for Theatrical Agents
Concept Maps Applied to Mars Exploration Public Outreach
Using Web Frequency Within Multimedia Exhibitions
The Association Engine: A Free Associative Digital Improviser
Between Now and the Semantic Web
Imagination Environment: Using the web as a source of popular culture
TextPool: Visualizing Live Text Streams
Keyless Media Delivery and Security: Location-aware computing through Interaction Design
Network Arts: Exposing Cultural Reality
Low-Fidelity Location Based Information Systems
Context-Aware Keyless Computing
Beyond Broadcast
Beyond Broadcast: a demo
Towards a Non-Linear Narrative Construction
Clustering for Opportunistic Communication
Automatically Indexing Documents: Content vs. Reference
Java Settlers: A research Environment for Studying Multi-Agent Negotiation
Supporting Online Resource Discovery in the Context of Ongoing Tasks with Proactive Software Assistants
Flytrap: Intelligent Group Music Recommendation
Using Citations to Facilitate Precise Indexing and Automatic Index Creation in Collections of Research Papers
XLibris: An Automated Library Research Assistant
Facilitating Opportunistic Communication by Tracking the Documents People Use
Information access in context
Beyond Similarity
Jabberwocky: You don't have to be a rocket scientist to change slides for a hydrogen combustion lecture
Guiding People to Information: Providing an Interface to a Digital Library Using Reference as a Basis for Indexing
Improving Human Computer Interaction in a Classroom Environment using Computer Vision
User Interactions with Everyday Applications as Context for Just-in-time Information Access
Mining Navigation History for Recommendation
Constructing Indices from Citations in Collections of Research Papers
CBR Textuality
Q&A: A System for the Capture, Organiation and Reuse of Expertise
Selecting Task-Relevant Sources for Just-in-Time Retrieval
Automatically Indexing Research Papers Using Text Surrounding Citations
InfoLab Package
Beyond "Next slide, please": The use of content and speech in multi-modal control
Learning for Question Answering and Text Classification: Integrating Knowledge-Based and Statistical Techniques
All gadget and no representation makes Jack a dull environment
Cooperating with people: The Intelligent Classroom
Gargoyle: Vision in the Intelligent Classroom
Integrating Range and Object Data for Robot Navigation
Happy Patrons Make Better Tippers: Creating a Robot Waiter Using Perseus and the Animate Agent Architecture
Machine-Generated Multimedia Content
From Generating to Mining: Automatically Scripting Conversations Using Existing Online Sources
News at Seven: The Future of the Future.
Classifying Paintings by Artistic Genre: An Analysis of Features & Classifiers
