Motivated by the growing practice of using social network data in credit scoring, we analyze impact network-based measures on customer score accuracy and tie formation among customers. We develop a series models to compare scores obtained with without data. also investigate how changes when consumers can strategically construct their networks attain higher scores. find that those who are motivated improve may form fewer ties focus more similar partners. The such endogenous consumer is...

We show that a two-sided matching platform can successfully compete by limiting the number of choices it offers to its customers, while charging higher prices than platforms with unrestricted choice. develop stylized model online dating where agents different outside options match based on how much they like each other. Starting from these microfoundations, we derive strength and direction indirect network effects increasing potential matches has positive effect due larger choice, but also...

This paper develops a theoretical model to study how the revenue and technology of social media platform affect its content-moderation strategy.

Imbalanced data is an important problem for medical analysis. Medical datasets are often not balanced in their class labels. The traditional classifiers can be seriously affected by the imbalanced distribution data. This because they aim to optimize overall accuracy without considering relative of each class. study searches effect imbalance training when developing neural network classifier decision making on chronic kidney disease. Neural networks widely used a number applications including...

Political campaigns are among the most sophisticated marketing exercises in United States. As part of their communication strategy, an increasing number politicians adopt social media to inform constituencies. This study documents returns from adopting a new technology, namely Twitter, for running Congress by focusing on change campaign contributions received. We compare weekly donations received just before and after politician opens Twitter account regions with high low levels penetration,...

How do social interactions shape collective action, and how are they mediated by networked information technologies? We answer these questions studying the Temperance Crusade, a wave of anti-liquor protest activity spreading across 29 states between 1873 1874. Relying on exogenous variation in network links generated railroad accidents, we provide causal evidence driving diffusion movement, rail telegraph about neighboring activity. Local newspaper coverage crusade was key channel mediating...

Using microdata from the American Red Cross (ARC) and Federal Election Commission (FEC) in two natural experiments, we provide evidence that political giving charitable are substitutes. In first experiment, estimate effects of a positive shock to donations ARC: foreign disaster events. We find although ARC increase by 34.9% six weeks following disaster, decline 18.8% same period. Put differently, each 1% is accompanied 0.53% drop donations. At average county-week–level donations, implied...

In this study, we investigate newspapers' decision to expand their product lines by adding online editions that incorporate user-generated content. We demonstrate such line extensions mitigate the extent of slanting in print media. The results also show as discretion users generate content increases, versions newspapers become more polarized. Furthermore, reduced profits for additional variants increase intensity competition market and awarded limits ability extract rents from consumers....

Abstract This paper studies how automation impacts the structure of decision‐making in organizations. We develop a theoretical model firm, where principal makes decision about much to prioritize new product development division when is led by manager who holds private information specific this and has misaligned preferences with principal. The chooses whether decentralize delegating it manager, resulting more informed but unbiased decision. In setting, we investigate which reduces...

Class imbalance problem is one of the important problems for classification studies in data mining. In this study, a comparative analysis some sampling methods was performed based on evaluation four algorithms prediction albendazole adverse events outcomes. Albendazole main medications used treatment variety parasitic worm infestations. The dataset created from public release FDA's FAERS database. Four were to analyze and their performance evaluated by using classifiers. Among algorithms,...

The MEDLINE database (Medical Literature Analysis and Retrieval System Online) contains an enormously increasing volume of biomedical articles. There is urgent need for techniques which enable the discovery, extraction, integration use hidden knowledge in those Text mining aims at developing technologies to help cope with interpretation these large volumes publications. Co-occurrence analysis a technique applied text methodologies statistical models are used evaluate significance...

Motivated by the growing practice of using social network data in credit scoring, this study analyzes impact based measures on customer score accuracy and tie formation among customers. We develop a series models to compare scores obtained with without data. also investigate how changes when individuals can strategically construct their networks attain higher scores. find that, if are motivated improve scores, they may form fewer ties focus them more similar partners. The such endogenous...

Download This Paper Open PDF in Browser Add to My Library Share: Permalink Using these links will ensure access this page indefinitely Copy URL DOI

Abstract Engineering programs must assess students' abilities to master “criteria 3 a‐k.” Skills such as teamwork, problem solving, design, and ethical understanding entail learning various processes; hence, assessing these outcomes is better accomplished by focusing on the process rather than result. Methods for observing performance, 100 percent behavioral observation, are ideal but expensive. We extend work sampling, an economic industry‐based alternative, observe cognitive processes....

This paper develops a theoretical model to study the economic incentives for social media platform moderate user-generated content. We show that self-interested can use content moderation as an effective marketing tool expand its installed user base, increase utility of users, and achieve positioning or extreme platform. The optimal strategy differs platforms with different revenue models, advertising subscription. also platform's depends on technical sophistication. Because imperfect...

We show that a two-sided platform can successfully compete by limiting the choice of potential matches it offers to its customers while charging higher prices than platforms with unrestricted choice. Starting from micro-foundations, we derive strength and direction network effect, find increasing number not only has positive effect due larger choice, but also negative competition between agents on same side. Agents heterogeneous outside options resolve trade-off two effects differently. For lower...