Zachary Shelton

In the past, researchers studied readability enhancement of English articles for non-native English readers, either on paper reading or hypertext reading. Using a variety of methods, researchers were able to enhance the reading comprehension and the users' satisfaction on hypertext reading, such as changing content presentation with visual-syntactic text formatting (VSTF) format or Jenga format. In terms of dynamically changing content presentation for reading, one less explored format is Portable Document Format (PDF), which was traditionally viewed within a modern Web browser or Adobe Acrobat reader on the desktop. PDF format was standardized as an open format in 2008 and has been widely used to keep a fixed-layout content. However, a fixed layout document presents a challenge to apply existing transformation methods, not mention on mobile devices. In this paper, we present a system that uses a novel algorithm to decode a PDF document and apply content transformation to enhance its readability. Although we used Jenga format as an example to enhance the readability of PDF documents, we envision the proposed framework can be used to adopt different transformation methods. The system was implemented in a mobile device and we are able to apply a basic transformation to a PDF document at both the sentence and paragraph levels. The main contribution of this research is we extend previous work of readability enhancement from paper document and hypertext content to PDF documents. Current result is promising, and we believe it is worth further investigation to make PDF documents readable and accessible on the Web for different populations, such as non-native English readers, people with dyslexia or special needs, etc.

Luxury is an Android application that approximates the current UV index by using data from both the front facing camera and the built in light sensor.

A machine learning implementation project that focuses on using facial recognition to find specified faces in crowds.

Forecast is an open sourced data forecasting program written in Java. It incorporates a robust plugin system that allows for custom file import schemas and user defined prediction algorithms. It all gets tied together with a JavaFX based UI that is easy to use and pleasant to look at.

Boston HistoryTrek is for history lovers, tourists, and anyone who wants to explore Historical Boston. Launch the app to uncover historical places, add locations to build your own itinerary, and launch navigation to start exploring!

Real-time Interactive Collaborative Editor (RICE) is a simple, easy to use text editor that you can use to collaboration with groups of people all shapes and sizes.