1
|
Kalir JH, Morales E, Fleerackers A, Alperin JP. “When I saw my peers annotating”. INFORMATION AND LEARNING SCIENCES 2020. [DOI: 10.1108/ils-12-2019-0128] [Citation(s) in RCA: 5] [Impact Index Per Article: 1.3] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/17/2022]
Abstract
Purpose
Social annotation (SA) is a genre of learning technology that enables the annotation of digital resources for information sharing, social interaction and knowledge production. This study aims to examine the perceived value of SA as contributing to learning in multiple undergraduate courses.
Design/methodology/approach
In total, 59 students in 3 upper-level undergraduate courses at a Canadian university participated in SA-enabled learning activities during the winter 2019 semester. A survey was administered to measure how SA contributed to students’ perceptions of learning and sense of community.
Findings
A majority of students reported that SA supported their learning despite differences in course subject, how SA was incorporated and encouraged and how widely SA was used during course activities. While findings of the perceived value of SA as contributing to the course community were mixed, students reported that peer annotations aided comprehension of course content, confirmation of ideas and engagement with diverse perspectives.
Research limitations/implications
Studies about the relationships among SA, learning and student perception should continue to engage learners from multiple courses and from multiple disciplines, with indicators of perception measured using reliable instrumentation.
Practical implications
Researchers and faculty should carefully consider how the technical, instructional and social aspects of SA may be used to enable course-specific, personal and peer-supported learning.
Originality/value
This study found a greater variance in how undergraduate students perceived SA as contributing to the course community. Most students also perceived their own and peer annotations as productively contributing to learning. This study offers a more complete view of social factors that affect how SA is perceived by undergraduate students.
Collapse
|
2
|
Abstract
The most commonly used algorithm in recommendation systems is collaborative filtering. However, despite its wide use, the prediction accuracy of this algorithm is unexceptional. Furthermore, whether quantitative data such as product rating or purchase history reflect users’ actual taste is questionable. In this article, we propose a method to utilise user review data extracted with opinion mining for product recommendation systems. To evaluate the proposed method, we perform product recommendation test on Amazon product data, with and without the additional opinion mining result on Amazon purchase review data. The performances of these two variants are compared by means of precision, recall, true positive recommendation (TPR) and false positive recommendation (FPR). In this comparison, a large improvement in prediction accuracy was observed when the opinion mining data were taken into account. Based on these results, we answer two main questions: ‘Why is collaborative filtering algorithm not effective?’ and ‘Do quantitative data such as product rating or purchase history reflect users’ actual tastes?’
Collapse
Affiliation(s)
- Youdong Yun
- Department of Computer Science and Engineering, Korea University, Korea
| | - Danial Hooshyar
- Department of Computer Science and Engineering, Korea University, Korea
| | - Jaechoon Jo
- Department of Computer Science and Engineering, Korea University, Korea
| | - Heuiseok Lim
- Department of Computer Science and Engineering, Korea University, Korea
| |
Collapse
|
3
|
Yun Y, Hooshyar D, Jo J, Lim H. Developing a hybrid collaborative filtering recommendation system with opinion mining on purchase review. J Inf Sci 2017. [DOI: 10.1177/0165551517692955] [Citation(s) in RCA: 29] [Impact Index Per Article: 4.1] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/16/2022]
Abstract
The most commonly used algorithm in recommendation systems is collaborative filtering. However, despite its wide use, the prediction accuracy of this algorithm is unexceptional. Furthermore, whether quantitative data such as product rating or purchase history reflect users’ actual taste is questionable. In this article, we propose a method to utilise user review data extracted with opinion mining for product recommendation systems. To evaluate the proposed method, we perform product recommendation test on Amazon product data, with and without the additional opinion mining result on Amazon purchase review data. The performances of these two variants are compared by means of precision, recall, true positive recommendation (TPR) and false positive recommendation (FPR). In this comparison, a large improvement in prediction accuracy was observed when the opinion mining data were taken into account. Based on these results, we answer two main questions: ‘Why is collaborative filtering algorithm not effective?’ and ‘Do quantitative data such as product rating or purchase history reflect users’ actual tastes?’
Collapse
Affiliation(s)
- Youdong Yun
- Department of Computer Science and Engineering, Korea University, Korea
| | - Danial Hooshyar
- Department of Computer Science and Engineering, Korea University, Korea
| | - Jaechoon Jo
- Department of Computer Science and Engineering, Korea University, Korea
| | - Heuiseok Lim
- Department of Computer Science and Engineering, Korea University, Korea
| |
Collapse
|
4
|
The Impacts of Mutual Collaboration Experience and Domain Knowledge Levels on CAIS Behavior: An Experimental Study. LIBRI 2017. [DOI: 10.1515/libri-2016-0061] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/15/2022]
Abstract
AbstractSome studies have investigated participants’ information-seeking practices from the perspective of mutual collaboration experience and level of domain knowledge. This study provides insights into collaborative academic information-seeking (CAIS) behaviour. The article explores whether groups with different levels of domain knowledge and mutual collaboration experiences had different CAIS behaviour. It also asked whether domain knowledge level or mutual collaboration experience had an impact on CAIS behaviour. We describe a user study with 18 participants in nine pairs with an experimental collaborative information-seeking tool, the participants categorized into three types of groups: high domain knowledge level and few mutual collaboration experiences (Group 1), low domain knowledge level and many mutual collaboration experiences (Group 2), and no domain knowledge level and no mutual collaboration experience (Group 3). Quantitative and qualitative data analysis were used to analyse the user data collected. The results showed that compared with members of Group 3, participants in Groups 1 and 2 had a better understanding of search tasks and were aware of the ways of completing the tasks successfully. They did not depend on the information-retrieval system when constructing search queries, and adopted diverse cooperation strategies. They were more likely to recommend information to their partners. Domain knowledge had greater impact on CAIS behaviour than collaboration experience. The findings help us to understand social interactions among community members and help CAIS researchers to understand user interactions and inform information system designers as they design collaborative systems to facilitate social communication in the information-seeking process. Our work was limited by the group types we chose and the small group size, which could affect the generalizability of our findings and should be addressed in future studies.
Collapse
|
5
|
Wang W, Wang H. Opinion-enhanced collaborative filtering for recommender systems through sentiment analysis. NEW REV HYPERMEDIA M 2015. [DOI: 10.1080/13614568.2015.1074726] [Citation(s) in RCA: 5] [Impact Index Per Article: 0.6] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/23/2022]
|
6
|
|
7
|
Du HS, Chu SK, Gorman GE, Siu FL. Academic social bookmarking: An empirical analysis of Connotea users. LIBRARY & INFORMATION SCIENCE RESEARCH 2014. [DOI: 10.1016/j.lisr.2013.06.003] [Citation(s) in RCA: 7] [Impact Index Per Article: 0.7] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/30/2022]
|