Abstract:
Online social media, a user-generated content or online word of mouth (WOM), which allows consumers to share their product opinions and experience and has the potential to influence product sales and firm strategy, is studied in context of the Hollywood movie industry. An online WOM information was collected from the message board of Yahoo Movies for a total of 257 movies released from 2005 to 2006. SentiWordNet and OpinionFinder, two lexical packages of computational linguistics, were used to construct the sentiment measures for the WOM data. Results show that WOM communication starts early in the preproduction period, becomes highly active before movie release, and diminishes as the movie is shown for more weeks in theaters. A movie that receives more active WOM communication tends to receive higher evaluations from movie critics, suggesting the number of messages could work as a signal for product quality.
Abstract:
Information overload often hinders knowledge discovery on the Web. Existing tools lack analysis and visualization capabilities. Search engine displays often overwhelm users with irrelevant information. This research proposes a visual framework for knowledge discovery on the Web. The framework incorporates Web mining, clustering, and visualization techniques to support effective exploration of knowledge. Two new browsing methods were developed and applied to the business intelligence domain: Web community uses a genetic algorithm to organize Web sites into a tree format; knowledge map uses a multidimensional scaling algorithm to place Web sites as points on a screen. Experimental results show that knowledge map out-performed Kartoo, a commercial search engine with graphical display, in terms of effectiveness and efficiency. Web community was found to be more effective, efficient, and usable than result list. Our visual framework thus helps to alleviate information overload on the Web and offers practical implications for search engine developers. © 2005 M.E. Sharpe, Inc.
Abstract:
Corporate credit rating analysis has attracted lots of research interests in the literature. Recent studies have shown that Artificial Intelligence (AI) methods achieved better performance than traditional statistical methods. This article introduces a relatively new machine learning technique, support vector machines (SVM), to the problem in attempt to provide a model with better explanatory power. We used backpropagation neural network (BNN) as a benchmark and obtained prediction accuracy around 80% for both BNN and SVM methods for the United States and Taiwan markets. However, only slight improvement of SVM was observed. Another direction of the research is to improve the interpretability of the AI-based models. We applied recent research results in neural network model interpretation and obtained relative importance of the input financial variables from the neural network models. Based on these results, we conducted a market comparative analysis on the differences of determining factors in the United States and Taiwan markets. © 2003 Elsevier B.V. All rights reserved.
