Yen-Ling Kuo

Community-based game design: experiments on social games for commonsense data collection

Games with A Purpose have successfully harvested information from web users. However, designing games that encourage sustainable and quality data contribution remains a great challenge. Given that many online communities have enjoyed active participation from a loyal following, this research explores how human computation games may benefit from rich interactions inherent in a community. We experimented by implementing two games for commonsense data collection on the leading social community platforms: the Rapport Game on Facebook and the Virtual Pet Game on PTT. In this paper, we present the choices of interaction mode and goal-oriented user model for building a community-based game. The data quality, collection efficiency, player retention, concept diversity, and game stability of both games are analyzed quantitatively from data collected since August/November 2008. Our findings should provide useful suggestions for designing community-based games in the future.

Resource-bounded crowd-sourcing of commonsense knowledge

Knowledge acquisition is the essential process of extracting and encoding knowledge, both domain specific and commonsense, to be used in intelligent systems. While many large knowledge bases have been constructed, none is close to complete. This paper presents an approach to improving a knowledge base efficiently under resource constraints. Using a guiding knowledge base, questions are generated from a weak form of similarity-based inference given the glossary mapping between two knowledge bases. The candidate questions are prioritized in terms of the concept coverage of the target knowledge. Experiments were conducted to find questions to grow the Chinese ConceptNet using the English ConceptNet as a guide. The results were evaluated by online users to verify that 94.17% of the questions and 85.77% of the answers are good. In addition, the answers collected in a six-week period showed consistent improvement to a 36.33% increase in concept coverage of the Chinese commonsense knowledge base against the English ConceptNet.

Planning for Reasoning with Multiple Common Sense Knowledge Bases

Intelligent user interfaces require common sense knowledge to bridge the gap between the functionality of applications and the user’s goals. While current reasoning methods have been used to provide contextual information for interface agents, the quality of their reasoning results is limited by the coverage of their underlying knowledge bases. This article presents reasoning composition, a planning-based approach to integrating reasoning methods from multiple common sense knowledge bases to answer queries. The reasoning results of one reasoning method are passed to other reasoning methods to form a reasoning chain to the target context of a query. By leveraging different weak reasoning methods, we are able to find answers to queries that cannot be directly answered by querying a single common sense knowledge base. By conducting experiments on ConceptNet and WordNet, we compare the reasoning results of reasoning composition, directly querying merged knowledge bases, and spreading activation. The results show an 11.03% improvement in coverage over directly querying merged knowledge bases and a 49.7% improvement in accuracy over spreading activation. Two case studies are presented, showing how reasoning composition can improve performance of retrieval in a video editing system and a dialogue assistant.

Capability modeling of knowledge-based agents for commonsense knowledge integration

Robust intelligent systems require commonsense knowledge. While significant progress has been made in building large commonsense knowledge bases, they are intrinsically incomplete. It is difficult to combine multiple knowledge bases due to their different choices of representation and inference mechanisms, thereby limiting users to one knowledge base and its reasonable methods for any specific task. This paper presents a multi-agent framework for commonsense knowledge integration, and proposes an approach to capability modeling of knowledge bases without a common ontology. The proposed capability model provides a general description of large heterogeneous knowledge bases, such that contents accessible by the knowledge-based agents may be matched up against specific requests. The concept correlation matrix of a knowledge base is transformed into a k-dimensional vector space using low-rank approximation for dimensionality reduction. Experiments are performed with the matchmaking mechanism for commonsense knowledge integration framework using the capability models of ConceptNet, WordNet, and Wikipedia. In the user study, the matchmaking results are compared with the ranked lists produced by online users to show that over 85% of them are accurate and have positive correlation with the user-produced ranked lists.

Utilizing descriptive documents for adaptive and reconfigurable M2M systems

Machine-to-Machine (M2M) has becomes an important research topic in the recent years. There are application choices for users, but these applications cannot exchange information with each other. Applications for different purposes cannot share the same devices, because the providers are different. Therefore, the users need an integrated framework that can easily configure devices and adapt to the user requirements. We propose three descriptive documents to facilitate the achievement of reconfiguration and adaptation. Results demonstrate our framework can seamlessly manage devices and promptly adapt to the environmental changes.

ACTraversal: ranking crowdsourced commonsense assertions and certifications

Building commonsense knowledge bases is a challenging undertaking. While we have witnessed the successful collection of large amounts of commonsense knowledge by either automatic text mining or games with a purpose (GWAP), such data are of limited precision. Verifying data is typically done with repetition, which works better for very large data sets. Our research proposes a novel approach to data verification by coupling multiple data collection methods. This paper presents ACTraversal, a graph traversal algorithm for ranking data collected from GWAP and text mining. Experiments on aggregating data from two GWAPs, i.e. Virtual Pets and Top10, with two text mining tools, i.e. SEAL and Google Distance, showed significant improvements.