The term “news analytics” covers the set of techniques, formulas, and statistics that are used to summarize and classify public sources of information. Metrics that assess analytics also form part of this set. In this Chapter I will describe various news analytics and their uses.
News analytics is a broad field, encompassing and related to information retrieval, machine learning, statistical learning theory, network theory, and collaborative filtering.
Examples of news analytics applications are reading and classifying financial information to determine market impact: for developing bullishness indexes and predicting volatility (Antweiler and Frank, 2004); reversals of news impact (Antweiler and Frank, 2005); the relation of news and message-board information (Das, Martinez-Jerez, and Tufano, 2005); the relevance of risk-related words in annual reports for predicting negative returns (Li, 2006); for sentiment extraction (Das and Chen, 2007); the impact of news stories on stock returns (Tetlock, 2007); determining the impact of optimism and pessimism in news on earnings (Tetlock, Saar-Tsechansky, and Macskassy, 2008); predicting volatility (Mitra, Mitra, and diBartolomeo, 2008), and predicting markets (Leinweber and Sisk, 2010 and this volume, Chapter 6).
We may think of news analytics at three levels: text, content, and context. The preceding applications are grounded in text. In other words (no pun intended), textbased applications exploit the visceral components ...