Use this URL to cite or link to this record in EThOS:
Title: Going beyond relevance : role of effort in information retrieval
Author: Verma, Manisha
ISNI:       0000 0004 7228 8912
Awarding Body: UCL (University College London)
Current Institution: University College London (University of London)
Date of Award: 2017
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
The primary focus of Information Retrieval (IR) systems has been to optimize for Relevance. Existing approaches to rank documents or evaluate IR systems does not account for “user effort”. Currently, judges only determine whether the information provided in a given document would satisfy the underlying information need in a query. The current mechanism of obtaining relevance judgments does not account for time and effort that an end user must put forth to consume its content. While a judge may spend a lot of time assessing a document, an impatient user may not devote the same amount of time and effort to consume its content. This problem is exacerbated on smaller devices like mobile. While on mobile or tablets, with limited interaction, users may not put in too much effort in finding information. This thesis characterizes and incorporates effort in Information Retrieval. Comparison of explicit and implicit relevance judgments across several datasets reveals that certain documents are marked relevant by the judges but are of low utility to an end user. Experiments indicate that document-level effort features can reliably predict the mismatch between dwell time and judging time of documents. Explicit and preference-based judgments were collected to determine which factors associated with effort agreed the most with user satisfaction. The ability to locate relevant information or findability was found to be in highest agreement with preference judgments. Findability judgments were also gathered to study the association of different annotator, query or document related properties with effort judgments. We also investigate how can existing systems be optimized for relevance and effort. Finally, we investigate the role of effort on smaller devices with the help of cost-benefit models.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available