Text classification is important to better understand online media. A major problem for creating accurate text classifiers using machine learning is small training sets due to the cost of annotating them. On this basis, we investigated how SVM and NBSVM text classifiers should be designed to achieve high accuracy and how the training sets should be sized to efficiently use annotation labor. We used a four-way repeated-measures full-factorial design of 32 design factor combinations. For each design factor combination 22 training set sizes were examined. These training sets were subsets of seven public text datasets. We study the statistical variance of accuracy estimates by randomly drawing new training sets, resulting in accuracy estimates for 98,560 different experimental runs. Our major contribution is a set of empirically evaluated guidelines for creating online media text classifiers using small training sets. We recommend uni- and bi-gram features as text representation, btc term weighting and a linear-kernel NBSVM. Our results suggest that high classification accuracy can be achieved using a manually annotated dataset of only 300 examples.
Corporate reputation is an economic asset and its accurate measurement is of increasing interest in practice and science. This measurement task is difficult because reputation depends on numerous factors and stakeholders. Traditional measurement approaches have focused on human ratings and surveys, which are costly, can be conducted only infrequently and emphasize financial aspects of a corporation. Nowadays, online media with comments related to products, services, and corporations provides an abundant source for measuring reputation more comprehensively. Against this backdrop, we propose an information retrieval approach to automatically collect reputation-related text content from online media and analyze this content by machine learning-based sentiment analysis. We contribute an ontology for identifying corporations and a unique dataset of online media texts labelled by corporations' reputation. Our approach achieves an overall accuracy of 84.4%. Our results help corporations to quickly identify their reputation from online media at low cost.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.