Stevelab, LLC is working on an app that will be released within the year. For planning purposes, the company needs to be able to predict how well the app will spread in its initial phase. Because the pilot launch will be limited to Hillsborough County, FL, there is a limited population. The marketing strategy is to sell the application to places where people gather to mingle and socialize, with a particular focus to the bar scene. Customers of the bar can use the app for free. In addition to our direct sales efforts, we can expect the bar owners to talk to other owners, users will talk to other people, and users will talk to other bar employees creating a web of additional growth. These components are very similar to an SIR model used to predict the spread of infection by dividing a population into 3 categories, Susceptible, Infected, and Recovered. Ratios are applied to show how each division interacts with the other two. (Nicho, 2010) Adding a vector as a means of transmission gives a second population to track divided into 3 similar categories. (Wei, Li, & Martcheva, 2007) In addition to reactions among these three, there are also constants to represent how the two populations interact. This study tests the viability of using an SIR vector model to predict sales of our new app. We conclude that there are good matches between the biological data points and the business data points making the use of an SIR model for this purpose plausible and a prediction can be made. Due to pervasive estimations in actual figures, the predictions are not expected to be extremely accurate at this time. However, we have the ability to directly monitor all of these points as they happen, meaning we can use the same model with increasing accuracy as business progresses, even as early as our first week.

For millions of people with swallowing disorders, preventing potentially deadly aspiration pneumonia requires following prescribed safe eating strategies. But adherence is poor, and caregivers' ability to encourage adherence is limited by the onerous and socially aversive need to monitoring another's eating. We have developed an early prototype for an intelligent assistant that monitors adherence and provides feedback to the patient, and tested monitoring precision with healthy subjects for one strategy called a \"chin tuck.\" Results indicate that adaptations of current generation machine vision and personal assistant technologies can effectively monitor chin tuck adherence, and suggest feasibility of a more general assistant that encourages adherence to a range of safe eating strategies.

For millions of people with swallowing disorders, preventing potentially deadly aspiration pneumonia requires following prescribed safe eating strategies. But adherence is poor, and caregivers’ ability to encourage adherence is limited by the onerous and socially aversive need to monitoring another’s eating. We have developed an early prototype for an intelligent assistant that monitors adherence and provides feedback to the patient, and tested monitoring precision with healthy subjects for one strategy called a “chin tuck.” Results indicate that adaptations of current generation machine vision and personal assistant technologies could effectively monitor chin tuck adherence, and suggest the feasibility of a more general assistant that encourages adherence to a wide range of safe eating strategies.

Measurement of healthcare quality in multiple care settings is a top priority. The ability to keep pace with the evolving standards of care to ensure we are measuring the right things at the right time is critical. The flexibility to add, update, and retire quality indicators easily and in tight timescales is essential. The Payment Information Calculation System (PICS), utilized in multiple national contracts for healthcare quality data collection and calculation, is facilitating the ability to respond. The design of the system enables increased efficiency by incorporating rules-driven technology. The tool provides non-developer subject matter experts the ability to define clinical data collection business rules in an XML format used by the PICS consumer to generate data collection user interfaces. This flexibility provides the ability to construct new data collection modules complete with data capture definitions, flow-logic (skip patterns) and enforce data collection formats and validate captured data (edits).

Business Intelligence (BI) is a set of methodologies, processes, architectures, and technologies that transform raw data into meaningful and useful information. BI enables more effective decision-making through strategic, tactical and operational insights. Our unique approach and patient linkage allows for improved information delivery, as well as increased access to information, improved ease of use, provision of a platform for collaboration and knowledge-sharing, and the ability to process and present reports from millions of rows of data in a matter of seconds.

Automated identification of tasks in email messages can be very useful to busy email users. What constitutes a task varies across individuals and must be learned for each user. However, training data for this purpose tends to be scarce. This paper addresses the lack of training data using domain-specific semantic features in document representation for reducing vocabulary mismatches and enhancing the discriminative power of trained classifiers when the number of training examples is relatively small.

Email clients were not designed to serve as a task management tools, but a high volume of task-relevant information in email leads many people to use email clients for this purpose. Such usage aggravates a user's experience of email overload and reduces productivity. Prior research systems have sought to address this problem by experimentally adding task management capabilities to email client software. RADAR (Reflective Agents with Distributed Adaptive Reasoning) takes a different approach in which a software agent acts like a trusted human assistant. Many RADAR components employ machine learning to improve their performance. Human participant studies showed a clear impact of learning on useI peIformance metrics.

Email client software is widely used for personal task management, a purpose for which it was not designed and is poorly suited. Past attempts to remedy the problem have focused on adding task management features to the client UI. RADAR uses an alternative approach modeled on a trusted human assistant who reads mail, identifies task-relevant message content, and helps manage and execute tasks. This paper describes the integration of diverse AI technologies and presents results from human evaluation studies comparing RADAR user performance to unaided COTS tool users and users partnered with a human assistant. As machine learning plays a central role in many system components, we also compare versions of RADAR with and without learning. Our tests show a clear advantage for learning-enabled RADAR over all other test conditions.