A System to Support Memory Recall by Creating Tag Clouds from Calendars, Twitter, and Photos
Harumi Murakami, Masaki Matsumoto, Sho Matsuura, Shuta Watanabe · International Journal of Service and Knowledge Management · 2022
Human memory is crucial for many daily activities.However, human memory is unreliable.Since existing memory aids such as diaries and notebooks need some effort for writing and accessing, we present an approach to capture and display memory cues from the user's daily activities, such as writing schedules, posting SNS messages, and taking photos.We developed a system that supports memory recall by creating tag clouds from the user's calendars and Twitter posts.The main feature of this research is using the number of photos taken by users to weight the tags to recall impressive events.This is based on the hypothesis that impressive events are found on the days when users took many photos.We evaluated out tag cloud generation algorithms and our implemented prototype system and obtained the following results for user memory recall: (1) Our tag cloud (using our weighting method with calendar and Twitter data) is superior to other tag clouds.(2) Sorting by time in the tag cloud outperforms other sorting methods.(3) Our implemented prototype system is useful.