skip to main content


Title: Children and search tools: Evaluation remains unclear
As children search the internet for materials, they o en turn to search engines that, unfortunately, o er children li le support as they formulate queries to initiate the search process or examine resources for relevance. While some solutions have been proposed to address this, inherent to this issue is the need to evaluate the e ectiveness of these solutions. We posit that the evaluation of the diverse aspects involved in the search process – from query suggestion generation to resource retrieval – requires a complex, multi-faceted approach that draws on evaluation methods utilized in human-computer interaction, information retrieval, natural language processing, education, and psychology.  more » « less
Award ID(s):
1763649
NSF-PAR ID:
10099389
Author(s) / Creator(s):
; ; ; ; ;
Date Published:
Journal Name:
KidRec Workshop co-located with ACM IDC 2019
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. Given the more widespread nature of natural language interfaces, it is increasingly important to understand who are accessing those interfaces, and how those interfaces are being used. In this paper, we explore spellchecking in the context of web search with children as the target audience. In particular, via a literature review we show that, while widely used, popular search tools are ill-designed for children. We then use spellcheckers as a case study to highlight the need for an interdisciplinary approach that brings together natural language processing, education, human-computer interaction to address a known information retrieval problem: query misspelling. We conclude that it is imperative that those for whom the interfaces are designed have a voice in the design process. 
    more » « less
  2. While society continues to be transformed by insights from processing big data, the increasing rate at which this data is gathered is making processing in private clusters obsolete. A vast amount of big data already resides in the cloud, and cloud infrastructures provide a scalable platform for both the computational and I/O needs of big data processing applications. Virtualization is used as a base technology in the cloud; however, existing virtual machine placement techniques do not consider data replication and I/O bottlenecks of the infrastructure, yielding sub-optimal data retrieval times. This paper targets efficient big data processing in the cloud and proposes novel virtual machine placement techniques, which minimize data retrieval time by considering data replication, storage performance, and network bandwidth. We first present an integer-programming based optimal virtual machine placement algorithm and then propose two low cost data- and energy-aware virtual machine placement heuristics. Our proposed heuristics are compared with optimal and existing algorithms through extensive evaluation. Experimental results provide strong indications for the superiority of our proposed solutions in both performance and energy, and clearly outline the importance of big data aware virtual machine placement for efficient processing of large datasets in the cloud. 
    more » « less
  3. People often have difficulty in expressing their information needs. Many times this results from a lack of clarity about the task at hand, or the way an information or search system works. In addition, people may not know what they do not know. The former is addressed by search systems by providing recommendations, whereas there are no good solutions for the latter problem. Even when a search system makes recommendations, they are limited to suggesting objects such as queries and documents only. They do not consider providing suggestions for strategies, people, or processes. This Perspective Paper addresses it by showing how to investigate the nature of the work a person is doing, predicting the potential problems they may encounter, and providing help to overcome those problems. This help could be an object such as a document or a query, a strategy, or a person. This whole process is referred to as Information Fostering. Beyond crafting a general-purpose recommender system, Information Fostering is the idea of providing proactive suggestions and help to information seekers. This could allow them avoid potential problems and capture promising opportunities from a search process before it is too late. The current paper presents this new perspective by outlining desired characteristics of an Information Fostering system, envisioning application scenarios, and proposing a set of potential methods for moving forward. Beyond these details, the primary purpose of this paper is to offer a new viewpoint that looks at the other side of the information seeking coin, by bringing together ideas from human-computer interaction, information retrieval, recommender systems, and education. 
    more » « less
  4. Language model pre-training has spurred a great deal of attention for tasks involving natural language understanding, and has been successfully applied to many downstream tasks with impressive results. Within information retrieval, many of these solutions are too costly to stand on their own, requiring multi-stage ranking architectures. Recent work has begun to consider how to “backport” salient aspects of these computationally expensive models to previous stages of the retrieval pipeline. One such instance is DeepCT, which uses BERT to re-weight term importance in a given context at the passage level. This process, which is computed offline, results in an augmented inverted index with re-weighted term frequency values. In this work,we conduct an investigation of query processing efficiency over DeepCT indexes. Using a number of candidate generation algorithms, we reveal how term re-weighting can impact query processing latency, and explore how DeepCT can be used as a static index pruning technique to accelerate query processing without harming search effectiveness. 
    more » « less
  5. Hagen, Matthias and (Ed.)
    Readability is a core component of information retrieval (IR) tools as the complexity of a resource directly affects its relevance: a resource is only of use if the user can comprehend it. Even so, the link between readability and IR is often overlooked. As a step towards advancing knowledge on the influence of readability on IR, we focus on Web search for children. We explore how traditional formulas–which are simple, efficient, and portable–fare when applied to estimating the readability of Web resources for children written in English. We then present a formula well-suited for readability estimation of child-friendly Web resources. Lastly, we empirically show that readability can sway children’s information access. Outcomes from this work reveal that: (i) for Web resources targeting children, a simple formula suffices as long as it considers contemporary terminology and audience requirements, and (ii) instead of turning to Flesch-Kincaid–a popular formula–the use of the “right” formula can shape Web search tools to best serve children. The work we present herein builds on three pillars: Audience, Application, and Expertise. It serves as a blueprint to place readability estimation methods that best apply to and inform IR applications serving varied audiences. 
    more » « less