Entity Recommendation for Everyday Digital Tasks

Abstract

Recommender systems can support everyday digital tasks by retrieving and recommending useful information contextually. This is becoming increasingly relevant in services and operating systems. Previous research often focuses on specific recommendation tasks with data captured from interactions with an individual application. The quality of recommendations is also often evaluated addressing only computational measures of accuracy, without investigating the usefulness of recommendations in realistic tasks. The aim of this work is to synthesize the research in this area through a novel approach by (1) demonstrating comprehensive digital activity monitoring, (2) introducing entity-based computing and interaction, and (3) investigating the previously overlooked usefulness of entity recommendations and their actual impact on user behavior in real tasks. The methodology exploits context from screen frames recorded every 2 seconds to recommend information entities related to the current task. We embodied this methodology in an interactive system and investigated the relevance and influence of the recommended entities in a study with participants resuming their real-world tasks after a 14-day monitoring phase. Results show that the recommendations allowed participants to find more relevant entities than in a control without the system. In addition, the recommended entities were also used in the actual tasks. In the discussion, we reflect on a research agenda for entity recommendation in context, revisiting comprehensive monitoring to include the physical world, considering entities as actionable recommendations, capturing drifting intent and routines, and considering explainability and transparency of recommendations, ethics, and ownership of data.

EntityBot: Supporting Everyday Digital Tasks with Entity Recommendations

“…In this demo, we introduce EntityBot [11] to the RecSys community. EntityBot supports everyday digital tasks by predicting the information needs of the users and providing them with the right information at the right time through interactive entity recommendations.…”

“…The posterior has a closed form solution and is employed to estimate the expected relevance of entities and contexts (Equations 1 and 3) which respectively are used to rank all entities (of different types: people, keywords, and applications) and contexts (with their corresponding linked documents) to be recommended to the user. [11]. Recommended entities are displayed within four rows, here with five items each: people, applications, documents, and topics.…”

“…In [11] we introduce EntityBot, a system that supports everyday digital tasks with cross-app rich entity recommendation. EntityBot builds an unsupervised model of a user's topical activities from digital activity monitoring data.…”

“…Figure1: Two states of EntityBot's user interface[11]. Recommended entities are displayed within four rows, here with five items each: people, applications, documents, and topics.…”

Building Chinese interdisciplinary research centers : the case of Tsinghua University

“…Entity Set Expansion (ESE) is a critical task aiming to identify new entities belonging to the same semantic class as a given set of seed entities [15,18,19,46]. ESE plays an important role for a wide range of user-tailored applications [13,21,43,44], such as recommendation systems [7,23,41], knowledge base population Previous exploratory works mainly made efforts towards finegrained expansion. ProbExpan [18] refines fine-grained semantic boundaries by mining in-batch hard negative samples.…”