This paper shares our work on building a machine learning system to categorize transactions for Intuit's QuickBooks product. Transaction categorization is challenging due to the complexity of accounting, the need for personalization, and the diversity of customers. We have broken down this monolithic problem into smaller pieces based on customers' life-cycle stages, and tailored solutions to address customer pain-points for each. Modern machine learning technologies such as deep neural networks, transfer learning, and few-shot learning are adopted to enable accurate transaction categorization. Furthermore our system learns user actions in real-time to provide relevant and in-time category recommendations. This in-session learning capability reduces user workload, improves customer experience, and helps to cultivate confidence.

Abstract This study explores an image‐based retrieval interface for drug information, focusing on usability for a specific population—seniors. Qualitative, task‐based interviews examined participants' health information behaviors and documented search strategies using an existing database ( www.drugs.com ) and a new prototype that uses similarity‐based clustering of pill images for retrieval. Twelve participants (aged 65 and older), reflecting a diversity of backgrounds and experience with Web‐based resources, located pill information using the interfaces and discussed navigational and other search preferences. Findings point to design features (e.g., image enlargement) that meet seniors' needs in the context of other health‐related information‐seeking strategies (e.g., contacting pharmacists).