Michael Siering

Detecting Fraudulent Behavior on Crowdfunding Platforms: The Role of Linguistic and Content-Based Cues in Static and Dynamic Contexts

Abstract Crowdfunding platforms offer founders the possibility to collect funding for project realization. With the advent of these platforms, the risk of fraud has risen. Fraudulent founders provide inaccurate information or pretend interest toward a project. Within this study, we propose deception detection support mechanisms to address this novel type of Internet fraud. We analyze a sample of fraudulent and nonfraudulent projects published at a leading crowdfunding platform. We examine whether the analysis of dynamic communication during the funding period is valuable for identifying fraudulent behavior—apart from analyzing only the static information related to the project. We investigate whether content-based cues and linguistic cues are valuable for fraud detection. The selection of cues and the subsequent feature engineering is based on theories in areas of communication, psychology, and computational linguistics. Our results should be helpful to the stakeholders of crowdfunding platforms and researchers of fraud detection.

Crowdfunding Success Factors: The Characteristics of Successfully Funded Projects on Crowdfunding Platforms

Crowdfunding platforms offer promising opportunities for project founders to publish their project ideas and to collect money in order to be able to realize them. Consequently, the question of what influences the successful funding of projects, i.e., reaching the target amount of money, is very important. Building upon media richness theory and the concept of reciprocity, we extend previous research in the field of crowdfunding success factors. We provide a comprehensive view on factors influencing crowdfunding success by both focusing on project-specific as well as founder-specific aspects. Analyzing a sample of projects of the crowdfunding platform kickstarter.com, we find that the project description, related images and videos as well as the question of whether the founder has previously backed other projects influence funding success. Interestingly, the question of whether the founder has previously created other projects has no significant influence. Our results are of high interest for the stakeholders on crowdfunding platforms.

Boom or "Ruin"--Does It Make a Difference? Using Text Mining and Sentiment Analysis to Support Intraday Investment Decisions

Investors have to deal with an increasing amount of information in order to make beneficial investment decisions. Thus, text mining is often applied to support the decision-making process by predicting the stock price impact of financial news. Recent research has shown that there exists a relation between news article sentiment and stock prices. However, this is not considered by previous text mining studies. In this paper, we develop a novel two-stage approach that connects text mining with sentiment analysis to predict the stock price impact of company-specific news. We find that the combination of text mining and sentiment analysis improves forecasting results. Additionally, a higher accuracy can be achieved by using finance-related word lists for sentiment analysis instead of a generic dictionary.

How to enable automated trading engines to cope with news-related liquidity shocks? Extracting signals from unstructured data

Abstract Financial markets are characterised by high levels of complexity and non-linearity. Information systems have often been applied to support investors by forecasting price changes in securities markets. In addition to the asset price, liquidity represents another financial variable that has a high relevance for investors because it constitutes a main determinant of total transaction costs. Previous research has shown that the level of liquidity is affected by the publication of corporate disclosures. To derive an optimal order execution strategy that minimises the transaction costs, investors as well as automated trading engines must be able to anticipate changes in the available market liquidity. However, there is no research on how to forecast the impact of corporate disclosures on market liquidity. Therefore, we propose an IT artefact that allows automated trading engines to appropriately react to news-related liquidity shocks. The system indicates whether the publication of a regulatory corporate disclosure will be followed by a positive liquidity shock, i.e., lower transaction costs compared to historical levels. Utilising text mining techniques, the content of the corporate disclosures is analysed to generate a trading signal. Furthermore, the trading signal is evaluated within a simulation-based use case that considers English and German corporate disclosures and is shown to be of economic value.

Investigating the Impact of Media Sentiment and Investor Attention on Financial Markets

Media sentiment has been shown to be related to stock returns. However, one prerequisite for this influence has not been taken into account yet: the question of whether investors actually pay attention to news and the related financial instruments. Within this study, we close this research gap by examining the interplay between media sentiment and investor attention. Thereby, we find that the positive impact of media sentiment on returns is increased when investor attention is high. Furthermore, we evaluate whether these variables can be used to forecast future market movements. Although our results reveal that the obtained forecasting accuracy cannot be achieved by chance, we conclude that further information has to be included in the forecasting model to obtain satisfying results.

A taxonomy of financial market manipulations: establishing trust and market integrity in the financialized economy through automated fraud detection

Financial market manipulations represent a major threat to trust and market integrity in capital markets. Manipulations contribute to mispricing, market imperfections and an increase in transaction costs for market participants and in costs of capital for issuers. Manipulations are facilitated by increased transaction velocity, speculative trading and abusive usage of new trading technologies, i.e., they are directly linked to financial sector changes that drive financialization. Research at the intersection of financialization and IS might support regulatory authorities and market operators in improving market surveillance and helping to detect fraudulent activities. However, confusing terminology is prevalent on financial markets with respect to different manipulation techniques and their characteristics, which hampers efficient fraud detection. Furthermore, recognizing manipulations is challenging given the large number of information sources and the vast number of trades occurring not least because of high-frequency traders. Therefore, automated market surveillance tools require a comprehensive taxonomy of financial market manipulations as a basis for appropriate configuration. Based on a cluster analysis of SEC litigation releases, a review of the latest market abuse regulation and academic studies, we develop a taxonomy of manipulations that structures and details existing manipulation techniques and reveals how these techniques differ along several dimensions. In a case study, we show how the taxonomy can be utilized to guide the development of appropriate decision support systems for fraud detection.

The Role of Misbehavior in Efficient Financial Markets: Implications for Financial Decision Support

The analysis of different data sources to support financial decision making has been a subject of research for several decades. While early approaches mostly focus on structured data, recent studies also take into account unstructured data. In this paper, we build upon these two research streams and explore potential benefits that can be achieved by combining both approaches. Therefore, we present an approach that integrates both data types. From a theoretical perspective, our research angle is based on two fundamental theories in Finance: while the Efficient Market Hypothesis states that capital markets are information efficient, Behavioral Finance theory stresses that market efficiency may be limited, e.g. due to irrational behavior of market participants or market barriers. While the two theories provide arguments for and against the functioning of our approach, we can illustrate its superiority compared to other approaches. The implications are discussed from a methodological and theoretical perspective.

The economics of stock touting during Internet-based pump and dump campaigns

Information systems have facilitated the increase in relevance of financial markets. Nevertheless, the rise of the Internet has eased information‐based financial market manipulations. In this study, we examine the phenomenon of stock touting during pump and dump campaigns, in which deceivers advertise stocks to profit from an increased price level. We observe that the positive prospects promised are not confirmed by corporate disclosures and financial news. Furthermore, manipulators select targeted financial instruments based on specific stock and company characteristics. Manipulators avoid signals of anomaly and prefer unknown stocks. We find that stock touting has a positive market impact but that it is followed by a large decline in stock price in the subsequent days, causing investors to lose substantial amounts of their investments. We consider the impact of information generation, information content, and information presentation on the corresponding market reaction. Interestingly, information generation influences the demand for the stock, but information content and information presentation drive the willingness to pay. Our results are highly relevant for Internet users, software vendors, and market surveillance authorities, as a deep understanding of such information‐based manipulations is necessary to develop appropriate countermeasures.

Disentangling consumer recommendations: Explaining and predicting airline recommendations based on online reviews

Abstract Consumer recommendations of products and services are important performance indicators for organizations to gain feedback on their offerings. Furthermore, they are important for prospective customers to learn from prior consumer experiences. In this study, we focus on user-generated content, in particular online reviews, to investigate which service aspects are evaluated by consumers and how these factors explain a consumers recommendation. Further, we investigate how recommendations can be predicted automatically based on such user-driven responses. We disentangle the recommendation decision by performing explanatory and predictive analyses focusing on a sample of airline reviews. We identify core and augmented service aspects expressed in the online review. We then show that service aspect-specific sentiment indicators drive the decision to recommend an airline and that these factors can be incorporated in a predictive model using data mining techniques. We also find that the business model of an airline being reviewed, whether low cost or full service, is also an applicable consideration. Our results are highly relevant for practitioners to analyze and act on consumer feedback in a prompt manner, along with the ability of gaining a deeper understanding of the service from multiple aspects. Also, potential travelers can benefit from this approach by getting an aggregated view on service quality.