SEEN: Structured Event Enhancement Network for Explainable Need Detection of Information Recall Assistance

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 5438 - 5451 December 7-11, 2022 ©2022 Association for Computational Linguistics SEEN: Structured Event Enhancement Network for Explainable Need Detection of Information Recall Assistance You-En Lin,1 An-Zi Yen,2 Hen-Hsen Huang,3 Hsin-Hsi Chen1 1 Department of Computer Science and Information Engineering, National Taiwan University, Taiwan 2 Department of Computer Science, National Yang Ming Chiao Tung University, Taiwan 3 Institute of Information Science, Academia Sinica, Taiwan yelin@nlg.csie.ntu.edu.tw, azyen@nycu.edu.tw, hhhuang@iis.sinica.edu.tw, hhchen@ntu.edu.tw Abstract When recalling life experiences, people often forget or confuse life events, which necessitates information recall services. Previous work on information recall focuses on providing such assistance reactively, i.e., by retrieving the life event of a given query. Proactively detecting the need for information recall services is rarely discussed. In this paper, we use a human- annotated life experience retelling dataset to detect the right time to trigger the informa- tion recall service. We propose a pilot model– structured event enhancement network (SEEN) that detects life event inconsistency, additional information in life events, and forgotten events. A fusing mechanism is also proposed to incor- porate event graphs of stories and enhance the textual representations. To explain the need detection results, SEEN simultaneously pro- vides support evidence by selecting the related nodes from the event graph. Experimental re- sults show that SEEN achieves promising per- formance in detecting information needs. In addition, the extracted evidence can be served as complementary information to remind users what events they may want to recall. 1 Introduction People have to deal with many events in their daily life. As time passes, they might forget details about their past experiences. Forgetting the exact name of people or places or things and mixing up life events is a common occurrence. This explains the importance of an information recall system that helps people bring to mind what they are trying to recall. We propose reactive and proactive service modes for an information recall system (Yen et al., 2021a). In reactive mode, users directly ask the system about their life events, whereas in proactive mode, the system attempts to automatically detect whether users need memory recall assistance and then provides the information they seek to recall. For reactive mode, studies have been done on vi- sual lifelog recall (Gurrin et al., 2016, 2017, 2019, 2020; Chu et al., 2019, 2020), which focuses on the construction of a multimodal retrieval model that enables users to search through photos using tex- tual queries. We propose an information recall sys- tem (Yen et al., 2021b) to answer questions about life experiences over a personal knowledge base. In contrast to reactively receiving users’ requests, proactive mode, which detects the right time to trigger the information recall service, is still little explored. In this paper, we further propose a pi- lot study to proactively detect the user’s need for information recall assistance. One common use case of memory recall assis- tance occurs in human conversation. To identify whether people have difficulties in recalling …

Content truncated to 500 words