Event-based reminder application for prospective memory

• Main Author: Ter, Pei Ru
• Other Authors: Miao Chun Yan
• Format: Final Year Project
• Language: English
• Published: 2016
• Subjects: DRNTU::Engineering
• Online Access: http://hdl.handle.net/10356/66910
• Institution: Nanyang Technological University

One of the many health concern that comes with aging in these days includes the decline of prospective memory when an individual gets older. Prospective memory is a component that plays an important role in an individual's everyday memory. Deteriorating and failures in prospective memory is something that cannot be neglected as it can affects one’s ability in performing the prospective memory task, leading to serious consequences, such as forgotten to take daily medicine which may be life threatening. Although the failure in performing the prospective memory task cannot be fully avoided, it is possible to implement ways to act as an aid to minimize the failures in performing the prospective memory task to prevent disastrous consequences. The purpose of this project is to improve on an existing reminder application to include the location-based reminder as many existing reminder applications have been developed to aid people’s performance of time-based prospective memory tasks, yet few, if any, of them are catered for location-based prospective memory tasks. In order to remind effectively for location-based of prospective memory tasks, three main concerns are to be addressed: 1. How many reminders should be given for a task, 2. How should the reminder be delivered, and 3. When the reminder should be fired. A literature review had been done by studying on several psychological research paper on how various factors affecting the prospective memory task performance. After which a computational model is proposed based on the theoretical basis drew from the literature review. The model employs a Fuzzy Cognitive Map to predict performance of prospective memory tasks of an individual by considering the performance influencing factors. Given the individual performance prediction generated by the computational model, personalized number of reminders can be determined. This addresses the concern of point 1. To address the concern of point 2, sound notification method are said to be more suitable to increase the context awareness due to the environment. Due to insufficient aid using geofence, effective delivery of reminders through user’s movement of an area and the geographic layout of the space are needed. This addressed the concern of point 3. The proposed computational model is implemented in an Android application. To end off, this report includes the screen shots of the reminder application of how is it being used. Recommendations for future improvements to this research and application is also elaborated towards the end of the report.