Lisa Gandy

Classifying paintings by artistic genre: An analysis of features & classifiers

This paper describes an approach to automatically classify digital pictures of paintings by artistic genre. While the task of artistic classification is often entrusted to human experts, recent advances in machine learning and multimedia feature extraction has made this task easier to automate. Automatic classification is useful for organizing large digital collections, for automatic artistic recommendation, and even for mobile capture and identification by consumers. Our evaluation uses variableresolution painting data gathered across Internet sources rather than solely using professional high-resolution data. Consequently, we believe this solution better addresses the task of classifying consumer-quality digital captures than other existing approaches. We include a comparison to existing feature extraction and classification methods as well as an analysis of our own approach across classifiers and feature vectors.

A Software Application for Mining and Presenting Relevant Cancer Clinical Trials per Cancer Mutation

ClinicalTrials.org is a popular portal which physicians use to find clinical trials for their patients. However, the current setup of ClinicalTrials.org makes it difficult for oncologists to locate clinical trials for patients based on mutational status. We present CTMine, a system that mines ClinicalTrials.org for clinical trials per cancer mutation and displays the trials in a user-friendly Web application. The system currently lists clinical trials for 6 common genes (ALK, BRAF, ERBB2, EGFR, KIT, and KRAS). The current machine learning model used to identify relevant clinical trials focusing on the above gene mutations had an average 88% precision/recall. As part of this analysis, we compared human versus machine and found that oncologists were unable to reach a consensus on whether a clinical trial mined by CTMine was “relevant” per gene mutation, a finding that highlights an important topic which deems future exploration.

CashTagNN: Using sentiment of tweets with CashTags to predict stock market prices

In this paper we discuss a system, CashTagNN, which uses the sentiment and subjectivity scores of tweets that include cashtags of two companies, Apple and Johnson and Johnson, to model stock market movement, and in particular predict opening and closing stock market prices. We demonstrate that by using only sentiment and subjectivity along with a neural network machine learning model we can predict the opening and closing prices of the two companies with high accuracy.

Synthesizer: Expediting synthesis studies from context-free data with natural language processing

Today’s low cost digital data provides unprecedented opportunities for scientific discovery from synthesis studies. For example, the medical field is revolutionizing patient care by creating personalized treatment plans based upon mining electronic medical records, imaging, and genomics data. Standardized annotations are essential to subsequent analyses for synthesis studies. However, accurately combining records from diverse studies requires tedious and error-prone human curation, posing a significant barrier to synthesis studies. We propose a novel natural language processing (NLP) algorithm, Synthesize, to merge data annotations automatically. Application to patient characteristics for diverse human cancers and ecological datasets demonstrates the accuracy of Synthesize in diverse scientific disciplines. This NLP approach is implemented in an open-source software package, Synthesizer. Synthesizer is a generalized, user-friendly system for error-free data merging.

Pivot: Automatically Offering Information and Services to Real-World Shoppers

Shoppers with an Internet-enabled computer have a wealth of product information available to them. By browsing to a variety of Websites, users can conduct searches and compare prices, read reviews, and learn more about a product. These sites are pivot points for a user; once they are at Amazon.coms landing page, for example, they can navigate outwards to a million different products. The idiom of browsing to a central page for a site and then navigating outwards is acceptable when browsing is convenient, with large displays and useful input devices. This process becomes inconvenient, however, when the user is out and about in the world. We have built a system, Pivot, that uses physical objects as pivot points for the user. Specifically, Pivot uses the 1-D barcodes present on every product to deliver powerful services and options to a user on his or her cellphone. These services are chosen to be most useful to a user in the moment and trying to make a purchase decision. This paper describes the motivations for the system, the system itself, its current real-world deployment, and our intended future work.

Synthesizer: Expediting synthesis studies from context-free data with information retrieval techniques

Scientists have unprecedented access to a wide variety of high-quality datasets. These datasets, which are often independently curated, commonly use unstructured spreadsheets to store their data. Standardized annotations are essential to perform synthesis studies across investigators, but are often not used in practice. Therefore, accurately combining records in spreadsheets from differing studies requires tedious and error-prone human curation. These efforts result in a significant time and cost barrier to synthesis research. We propose an information retrieval inspired algorithm, Synthesize, that merges unstructured data automatically based on both column labels and values. Application of the Synthesize algorithm to cancer and ecological datasets had high accuracy (on the order of 85–100%). We further implement Synthesize in an open source web application, Synthesizer (https://github.com/lisagandy/synthesizer). The software accepts input as spreadsheets in comma separated value (CSV) format, visualizes the merged data, and outputs the results as a new spreadsheet. Synthesizer includes an easy to use graphical user interface, which enables the user to finish combining data and obtain perfect accuracy. Future work will allow detection of units to automatically merge continuous data and application of the algorithm to other data formats, including databases.

Innovations in News Media: Crisis Classification System

Research in crisis management is a relatively new area of study, originating in the 1980s. Researchers have created several different models that separate organizational crises into discrete stages, such as pre-crisis, crisis and post-crisis. In this article we discuss a natural language based crisis detection system which classifies news articles relating to crises into the appropriate crisis stage. We use news articles from the New York Times as a source of training data, and use this data along with state of the art data mining and machine learning algorithms as the core of the system. In the future, our system may be expanded to identify and evaluate crisis management strategies, suggest crisis management strategies for the current state of a crisis, or provide stakeholders with summaries of crises in news media.