Gender differentials and implicit feedback on online video content: enhancing user interest evaluation
| DOI | https://doi.org/10.1108/IMDS-12-2018-0551 |
| Date | 10 June 2019 |
| Pages | 1128-1146 |
| Published date | 10 June 2019 |
| Author | Woonkian Chong,Simon Rudkin,Junhui Zhang |
| Subject Matter | Information & knowledge management,Information systems,Data management systems,Knowledge management,Knowledge sharing,Management science & operations,Supply chain management,Supply chain information systems,Logistics,Quality management/systems |
Gender differentials and implicit
feedback on online video content:
enhancing user interest evaluation
Woonkian Chong
International Business School Suzhou,
Xi'an Jiaotong-Liverpool University, Suzhou, China
Simon Rudkin
School of Management, Swansea University, Swansea, UK, and
Junhui Zhang
Guangdong University of Science and Technology, Dongguan, China
Abstract
Purpose –Exponential growth in online video content makes viewing choice and video promotion
increasingly challenging. While explicit recommendation systems have value, they inherently distract the
user from normal behaviour and are open to numerous biases. To enhance user interest evaluation accuracy,
the purpose of this paper is to comprehensively examine the relationship between implicit feedback and
online video content, and reviews gender differentials in the interest indicated by a comprehensive set of
viewer responses.
Design/methodology/approach –This paper includes 200 useable observations based on an experiment
of user interaction with the Youku platform (one of the largest video-hosting websites in China). Logistic
regression was employed for its simple interpretation to test the proposed hypotheses.
Findings –The findings demonstrate gender differentials in cursor movement behaviour, explainable via
well-studied splits in personality, biological factors, primitive behaviour and emotion management. This work
offers a solution to the sparsity of work on implicit feedback, contributing to the literature that combines
explicit and implicit feedback.
Practical implications –This study offers a launch point for further work on human–computer
interaction, and highlights the importance of looking beyond individual metrics to embrace wider human
traits in video site design and implementation.
Originality/value –This paper links implicit feedback to online video content for the first time, and
demonstrates its value as an interest capturing tool. By reviewing gender differentials in the interest indicated
by a comprehensive set of viewer responses, this paper indicates how user characteristics remain critical.
Consequently, this work signposts highly fruitful directions for both practitioners and researchers.
Keywords Online video, User experience, Gender differentials, Implicit feedback, User interest
Paper type Research paper
1. Introduction
Online video delivers content on demand to inform and entertain an ever-growing number of
viewers, with improved internet accessibility further extending the reach of these videos.
A near unlimited bank of videos exists for every niche market, and that bank is
continuously increasing (Shehu et al., 2016). However, attempts to understand the appeal of
this content to its viewers predominantly focus on the reviews provided by individuals who
take the time to supply comments. Inherently, this divests the opinion from the viewing, and
produces many biases (Claypool et al., 2001). Implicit feedback attains the opinion of viewers
as they view digital content, thereby accurately collecting reaction data for conversion into
opinions (Tian et al., 2015). This paper contributes to the process of converting implicit data
into the interest measure that is needed by content providers, host websites and other video
users to inform their future designs, site contents and viewing choices. It delivers this
through a comprehensive analysis of implicit feedback measures, evaluating the power of
these measures as indicators of interest. This paper also reviews the role of gender as a
Industrial Management & Data
Systems
Vol. 119 No. 5, 2019
pp. 1128-1146
© Emerald PublishingLimited
0263-5577
DOI 10.1108/IMDS-12-2018-0551
Received 11 December 2018
Revised 26 March 2019
Accepted 26 March 2019
The current issue and full text archive of this journal is available on Emerald Insight at:
www.emeraldinsight.com/0263-5577.htm
1128
IMDS
119,5
mediator in the evaluation (Baker et al., 2007; Liu and Huang, 2008; Liu, 2016), as this is a
predominantly ignored area, yet an area that is essential if the captured implicit feedback is
to be correctly understood.
To understand the size of the industry, it is worth noting that, as of July 2015, 400 h of
content were uploaded to the video service YouTube every minute (Statista, 2018a, b). Such
a video information overload (Jones et al., 2004) makes it very easy for users tobecome lost,
bored and unable to find content that interests them (Browne et al., 2007; Zeng et al., 2012).
Subsequently, behaviour becomes sticky, slower to respond and less reflective of interests.
Recommendation systems predict user interest from explicit feedback, such as other user
reviews or user ratings (Claypool et al., 2001; Jawaheer et al., 2014; Akuma et al., 2016). Some
experts have further argued that this feedback can help improve recommendations
(Capuano et al., 2014). However, as Claypool et al. (2001) noted, the provision of feedback
diverts from normal web use and can feel cognitively burdensome (White et al., 2006),
possibly engendering rash choices or even refusal of related requests (Avery and
Zeckhauser, 1997). The resulting explicit feedback recommendations are read by other
users, yet contain many, or all, of these drawbacks. Ultimately, deficient recommendation
systems can drive users away (Montgomery, 2001). Thus, this paper seeks to remedy these
dual concerns of video overload and reviewer burden.
Implicit feedback notes users’actions and infers their recommendations based these
actions. Not requiring action from users aids compliance, while well-researched measures
validate the responses (Martin and Morich, 2011; Núñez-Valdéz et al., 2012; Balakrishnan
and Zhang, 2014). Moreover, individuals’reactions to things they do or do not like share
broad similarities, even when the videos are on very different topics. Although implicit
feedback has great potential for analyses, it cannot operate in isolation, and research must –
as does this paper –recognise the fundamental differences in the way different people react
to different stimuli, and attribute many of these differences to gender (Young et al., 1998;
Moe and Fader, 2004a, b; Danaher et al., 2006; Trauth, 2013; Li and Kannan, 2014;
Mallapragada et al., 2016). Although online media platforms, such as YouTube and Netflix,
hold a wealth of data and offer great scope for user-submitted opinion mining, there is still a
major shortcoming in the understanding of gender differentials in behaviour when viewing
online video content (Liu and Huang, 2008). It is crucial to examine gender differences in
computer use, especially in the process of user interface design and recommending the right
video content, to increase overall user experience. To provide a personalised and user-
friendly online video website with efficient recommendation systems, it is critical to
understand the nature and biological basis of gender differences in computer use (Yamauchi
et al., 2015), which is crucial for segmentation strategies (Teso et al., 2018). The current
gender differences literature (Imhof et al., 2007; Chai et al., 2016; Huang et al., 2016) divides
the focus into personality (Osorio et al., 2003), biological factors (Richard et al., 2007),
primitive behaviour (Meyers-Levy and Loken, 2015) and mood management (Meyers-Levy
and Loken, 2015; Yamauchi et al., 2015; Hibbeln et al., 2017). This paper reflects this critical
congruence in studying the implicit feedback of males and females.
Whether through viewer ratings, user reviews or surveys, explicit feedback delivers
deeper textual information, but is harder to extract meaning from. Provision of explicit
feedback alters the normal ways of understanding, reading or viewing online video because
of the interruptions/noises (Claypool et al., 2001). Thus, explicit feedback for online video
may be insufficient to capture user interest. In turn, this will influence the performance of
recommendation algorithms, such as collaborative filtering, which plays a vital role in
recommendation systems (Konstan et al., 1997). Thus, implicit feedback is considered a
promising alternative to explicit feedback (Balakrishnan and Zhang, 2014), especially for
video recommendation systems. A huge amount of implicit data can be captured from users’
natural interactions with online video websites. Recent research has started to incorporate
1129
Gender
differentials
and implicit
feedback
To continue reading
Request your trial