Social and Behavioral Sciences Commons^™

Monitoring Airport Service Quality: A Complementary Approach To Measure Perceived Service Quality Using Online Reviews, Kiljae Lee, Chunyan Yu

Chunyan Yu

Based on 42,063 airport reviews collected from Google Maps, we conducted a sentiment analysis and a topic modeling. We showed that the sentiment scores computed from textual reviews are good estimates of their paired star-ratings (r=0.63, p<0.01). Next, using the LDA (Latent Dirichlet Allocation), we extracted latent topics from the textual reviews and compared them with the standard categories utilized in the Airport Service Quality survey (ASQ). The topics extracted from reviews correspond well with the categories used in ASQ. We, in turn, compared the online ratings with the ratings annually updated by ASQ. While online reviews discuss almost identical topics with those of ASQ, the correlation between the ratings from two was weak (r=0.2). We suggest that the text mining approach using online reviews not only provides an inexpensive, dynamic, and locally customizable means of monitoring airport quality but also complements the standard survey by offering an alternative metric. ><0.01). Next, using the LDA (Latent Dirichlet Allocation), we extracted latent topics from the textual reviews and compared them with the standard categories utilized in the Airport Service Quality survey (ASQ). The topics extracted from reviews correspond well with the categories used in ASQ. We, in turn, compared the online ratings with the ratings annually updated by ASQ. While online reviews discuss almost identical topics with those of ASQ, the correlation between the ratings from two was weak (r=0.2). We suggest that the text mining approach using online reviews not only provides an inexpensive, dynamic, and locally customizable means of monitoring airport quality but also complements the standard survey by offering an alternative metric.

Citational Politics: Quantifying Impact In Digital Scholarship In The Humanities., Roopika Risam, Amy Earhart

Roopika Risam

Using Textual Features To Predict Popular Content On Digg, Paul H. Miller

Paul H Miller

Over the past few years, collaborative rating sites, such as Netflix, Digg and Stumble, have become increasingly prevalent sites for users to find trending content. I used various data mining techniques to study Digg, a social news site, to examine the influence of content on popularity. What influence does content have on popularity, and what influence does content have on users’ decisions? Overwhelmingly, prior studies have consistently shown that predicting popularity based on content is difficult and maybe even inherently impossible. The same submission can have multiple outcomes and content neither determines popularity, nor individual user decisions. My results show …

Assessment Of Brazil’S Research Literature, David J. Schoeneck, Alan L. Porter, Ronald N. Kostoff, Elena M. Berger

alan l porter

This “country study” analyzes substantial samples of research papers by Brazilian authors drawn from two global databases. The approach and the findings may each be of interest. Our approach is to examine R&D outputs through bibliometrics (to identify key authors, institutions, journals, etc) and text mining with taxonomy generation (to identify pervasive research thrusts). We extend prior country studies by providing for interactive data access and exploring military-relevant R&D information. The resulting publication activity profiles provide insight on Brazilian R&D strengths and investment strategies, and help identify opportunities for collaboration. Brazil, a nation of 190 million, evidences a substantial research …

The Use Of Environmental, Health And Safety Research In Nanotechnology Research, Jan Youtie, Alan Porter, Philip Shapira, Li Tang, Troy Benn

Li Tang

Environmental, health, and safety (EHS) concerns are receiving considerable attention in nanoscience and nanotechnology (nano) research and development (R&D). Policymakers and others have urged that research on nano's EHS implications be developed alongside scientific research in the nano domain rather than subsequent to applications. This concurrent perspective suggests the importance of early understanding and measurement of the diffusion of nano EHS research. The paper examines the diffusion of nano EHS publications, defined through a set of search terms, into the broader nano domain using a global nanotechnology R&D database developed at Georgia Tech. The results indicate that nano EHS research …