An Intra and Inter-Topic Evaluation and Cleansing Method
Journal Title: Romanian Journal of Human - Computer Interaction - Year 2010, Vol 3, Issue 2
Abstract
Topic modeling is a growing research field and novel ways of interpreting and evaluating results are necessary. We propose a method for evaluating and improving the performance of topic models generating algorithms relying on WordNet data. We first propose a measure for determining a topic model’s fitness factoring in its broadness and redundancy. Then, for each individual topic, the amount of relevant information it provides, along with its most important words and related concepts are determined by defining a cohesion function based on the topic’s projection on WordNet concepts. The model as a whole is improved by eliminating each topic’s outliers with respect to the ontology projection. We define a inter topic ontology based distance and we further use it to investigate the impact of removing redundant topics from a model with regard to the overlap between topics’ ontological projections. Clustering similar topics into conceptually cohesive groups is tried as an alternative to pruning less relevant topics. Results show that evaluating and improving statistical models with WordNet is a promising research track that leads to more coherent topic models.
Authors and Affiliations
Claudiu Muşat, Marian-Andrei Rizoiu , Ştefan Trauşan-Matu
Keywords
Related Articles
- EP ID EP28813
- DOI -
- Views 404
- Downloads 10
User Experience in Windows 7
The experience of a user with an operating system has the power to transform the software in a success or a failure. The new Windows version uses the test program feedback and the latest technologies, like multi-touch, i...
Prototypes Of Human-Machine Interactions And Types Of Educational Applications Specific To Mobile Augmented Reality
The technology of Augmented Reality - AR stimulates the perception of the surrounding reality by means of innovative and interactive human-machine interfaces. The development in an exponential rythm of the capabilities...
Evaluation and Automatic Summarization of Chat Conversations
With the continuous evolution of collaborative environments, the need for automatic analysis and participant assessment in Instant Messenger discussions (chats) has become more and more important. Moreover, a key element...
An Intra and Inter-Topic Evaluation and Cleansing Method
Topic modeling is a growing research field and novel ways of interpreting and evaluating results are necessary. We propose a method for evaluating and improving the performance of topic models generating algorithms relyi...
Recovering implicit thread structure in chat conversations
The analysis of chat conversations is a cumbersome task because of the number of different discussion threads that may occur at a certain moment. While most participants in a chat session tend to discuss one topic at a t...