Content Recommendation Services Research Articles

Equality of Learning Opportunity via Individual Fairness in Personalized Recommendations

Online education platforms play an increasingly important role in mediating the success of individuals’ careers. Therefore, while building overlying content recommendation services, it becomes essential to guarantee that learners are provided with equal recommended learning opportunities, according to the platform principles, context, and pedagogy. Though the importance of ensuring equality of learning opportunities has been well investigated in traditional institutions, how this equality can be operationalized in online learning ecosystems through recommender systems is still under-explored. In this paper, we shape a blueprint of the decisions and processes to be considered in the context of equality of recommended learning opportunities, based on principles that need to be empirically-validated (no evaluation with live learners has been performed). To this end, we first provide a formalization of educational principles that model recommendations’ learning properties, and a novel fairness metric that combines them to monitor the equality of recommended learning opportunities among learners. Then, we envision a scenario wherein an educational platform should be arranged in such a way that the generated recommendations meet each principle to a certain degree for all learners, constrained to their individual preferences. Under this view, we explore the learning opportunities provided by recommender systems in a course platform, uncovering systematic inequalities. To reduce this effect, we propose a novel post-processing approach that balances personalization and equality of recommended opportunities. Experiments show that our approach leads to higher equality, with a negligible loss in personalization. This paper provides a theoretical foundation for future studies of learners’ preferences and limits concerning the equality of recommended learning opportunities.

An AR mobile navigation system integrating indoor positioning and content recommendation services

Over the past few years, with the prevalence of smart mobile devices, several well-known museums and researchers have developed mobile navigation systems to improve the navigation effect for museum visitors. However, most of these systems lack efficient content recommendation, indoor positioning, and augmented reality (AR) display functions, and this situation causes difficulty for visitors to have a personalized and efficient navigation experience. This paper presents an augmented reality mobile navigation system that supports indoor positioning and content recommendation services, which aims to provide personalized suggestions corresponding to the user’s context through the analysis of the interest and location information of visitors. Moreover, combined with the developed markerless augmented reality display functions, this system could effectively enrich and improve visitor’s navigation experience.

다문화 구성원을 위한 학습자원 메타데이터 응용표준 프로파일

한국은 최근 급속도로 다문화 사회로 변모해 가고 있으며, 2015년 기준 다문화 가족수는 전체 인구의 3.5%, 80만명을 넘어서고 있다. 또한, 2016년 기준 국제결혼비율이 10%를 넘어서고 있어 다문화 가족수는 꾸준히 늘어날 전망이다. 이 연구는 한국 사회에 적응해야 하는 결혼이주여성들과 다문화 가족 구성원들의 필요와 선호에 맞춘 학습자원과 콘텐츠를 제공하기 위한 기반 조성의 일환으로 메타데이터 표준 프로파일을 설계한 것이다. 연구 추진 필요성을 검증하기 위해 소비자 집단을 선별하여 심층인터뷰를 가졌으며, 메타데이터 표준 프로파일 설계를 위해 공적 표준으로 채택된 관련 국제표준과 한국의 국가표준 시리즈를 분석하였다. 이어서 다문화 구성원을 위한 콘텐츠 특성을 분석하여 필요한 메타데이터 요소들을 선별하여 프로파일로 구성하였으며, 콘텐츠 제작자들의 요구를 반영하여 필수와 선택 조건들을 정의하였다. 본 연구는 결혼이주여성들의 교육 수요를 분석하여, 한-한 변환 시스템, 맞춤형 학습콘텐츠 추천 서비스, 학습관리시스템(Learning Management System) 등 효과적인 교육 콘텐츠를 개발하고 서비스하기 위해 필요한 메타데이터 표준안으로 기반을 조성했다는 점과 향후 교육과정의 주제별 서비스를 위해서도 자원이 노출될 수 있다는 점에서 의의가 있다. Korea is rapidly becoming a multicultural society in recent years, and the number of multicultural families in 2015 exceeds 3.5% and 800,000. Also, as international marriage rate exceeds 10% by 2016, the number of multicultural families is expected to steadily increase. This study is a design of a metadata application profile as part of the foundation for providing learning resources and content tailored to the needs and preferences of married immigrant women and multicultural family members who need to adapt to Korean society. In order to verify the necessity of the research, we conducted an in-depth interview by screening consumer groups, and analyzed the relevant international and Korean national standards as de-jure standards for the design of metadata standard profiles. Then, we analyzed the contents characteristics for multicultural members, and organized the necessary metadata elements into profiles. We defined the mandatory/optional conditions to reflect the needs of content providers. This study is meaningful in that the study analyzes the educational needs of married immigrant women and presents the necessary metadata standards to develop and service effective educational content, such as korean-to-korean conversion system, personalized learning contents recommendation service, and learning management system.

A comparative study of user intention to recommend content on mobile social networks

This study aims to explore user intention to recommend multimedia content on mobile social networks. To better understand user behavioral differences in content recommendations, this study utilizes user behavioral responses on social network services to determine heavy and light users. By analyzing data collected from 258 respondents, the findings reveal that the factors that influence intention to recommend vary among heavy and light users. First, trust, subjective norm, perceived ease of use, and perceived usefulness are considered as predictors for heavy users. Second, subjective norm, trust, perceived ease of use, and perceived usefulness are not influencing factors relative to recommendation intention for light users. Third, trust facilitates heavy users to share their content recommendations on mobile social networks. From theoretical perspectives, the results confirm that dynamic trust transfer could be integrated using the theory of planned behavior with a technology acceptance model. Considering practical implications, our findings regarding the prediction of heavy users provide business insights to content recommendation service. Our study highlights trust strategies related to migrating light users to heavy users. Overall, mobile social network providers must consider user technology perception enhancements and reduce trust concerns. Our findings contribute to theoretical applications and provide practical implications for social service providers in relation to social applications on mobile devices.

Optimization in Online Content Recommendation Services: Beyond Click-Through Rates

A new class of online services allows Internet media sites to direct users from articles they are currently reading to other content they may be interested in. This process creates a “browsing path” along which there is potential for repeated interaction between the user and the provider, giving rise to a dynamic optimization problem. A key metric that often underlies this recommendation process is the click-through rate (CTR) of candidate articles. Whereas CTR is a measure of instantaneous click likelihood, we analyze the performance improvement that one may achieve by some lookahead that accounts for the potential future path of users. To that end, by using some data of user path history at major media sites, we introduce and derive a representation of content along two key dimensions: clickability, the likelihood to click to an article when it is recommended; and engageability, the likelihood to click from an article when it hosts a recommendation. We then propose a class of heuristics that leverage both clickability and engageability, and provide theoretical support for favoring such path-focused heuristics over myopic heuristics that focus only on clickability (no lookahead). We conduct a live pilot experiment that measures the performance of a practical proxy of our proposed class, when integrated into the operating system of a worldwide leading provider of content recommendations, allowing us to estimate the aggregate improvement in clicks per visit relative to the CTR-driven current practice. The documented improvement highlights the importance and the practicality of efficiently incorporating the future path of users in real time.

Introduction to the Special Issue on Practice-Focused Research

When Steve Graves first approached us with the idea of serving as guest editors-in-chief for a special issue on practice-focused research, the appeal of this idea was immediate. We were very excited by this opportunity to highlight and promote the crucial role practice-focused research fulfills in our profession, and in society as a whole. At its core operations management (OM) concerns making life better: reducing waste, doing more with less, delivering services that meet people’s needs, and answering fundamental questions about how our society, and the business and service constructs we have created, operate. The papers in this special issue provide excellent examples of all of these characteristics. As we, and Steve, discuss in the forum piece at the start of the special issue, “Practice-Based Research in Operations Management: What It Is, Why Do It, Related Challenges, and How to Overcome Them” (Gallien et al. 2016), there are different strategies OM researchers can use to guide their research activities. Sometimes an OM researcher attains results through the use of mathematical constructs, which have the benefit of great generality and flexibility due to their abstraction of reality. Unfortunately, research of this type may be too abstract, and thus demonstrate limited validity in practice. Other times an OM researcher may directly address an extant problem in industry or society. One potential drawback of this latter research strategy is that such work may be too specific, limited in its application to the idiosyncratic problem context, and possibly other very closely related settings. What we sought for this special issue were those select few papers that captured the best of both of these streams of research; this coupling of generality born from abstraction and validity born from application became the defining principle we embraced about practice-based research. This principle is common to all of the papers in the special issue: They both utilize analytical sophistication to develop general insights and results, and yet also are able to establish their validity and relevance to practice. This is done through an application that yields verifiable and quantifiable results, or through an examination of a business phenomenon that again yields verifiable and quantifiable insights. We are very pleased to be able to present to you a set of nine papers that exemplify these goals, spanning a variety of industry sectors. The first three papers address novel manifestations of classical OM concerns, all within the swiftly evolving technology sector. The first of these, “Optimization in Online Content Recommendation Services: Beyond Click-Through Rates” by Besbes et al. (2016), addresses the fundamental question of which products to offer to customers. But when evaluating the products they can choose from—content recommendation (links) at the end of Web pages—they recognize the need to consider not only whether the product will be appealing to the customer, i.e., whether the content will entice the reader to click on it (“clickability”) but also whether the product will be likely to entice the reader to consume a subsequent product, i.e., whether it is likely the reader will click from it to yet another article (“engageability”). The authors show that by combining these two concepts the algorithm they develop can arrive at superior recommendations. This paper provides an admirable combination of analytical work deriving and then demonstrating the near-optimality of their algorithm, and validation work in a controlled pilot with their industry partner, Outbrain, illustrating when consideration of engageability can be most valuable. The second paper within this set, “Proactive Customer Education, Customer Retention, and Demand

MSOM Society Student Paper Competition: Abstracts of 2014 Winners

The journal is pleased to publish the abstracts of the six finalists of the 2014 Manufacturing and Service Operations Management Society’s student paper competition. The 2014 prize committee was chaired by Göker Aydın (Kelley School of Business, Indiana University), Guillaume Roels (UCLA Anderson School of Management, University of California, Los Angeles), and Gilvan Souza (Kelley School of Business, Indiana University). The other committee members were Philipp Afèche, Vishal Agrawal, Gad Allon, Aydın Alptekinoğlu, Alessandro Arlotto, Mor Armony, Atalay Atasu, Opher Baron, Bob Batt, Kostas Bimpikis, Robert Bray, René Caldentey, Carri Chan, Li Chen, Xin Chen, Ying-Ju Chen, Soo-Haeng Cho, So Yeon Chun, Nicole DeHoratius, Sarang Deo, Lingxiu Dong, David Drake, Pnina Feldman, Santiago Gallino, Srinagesh Gavirneni, Karan Girotra, Manu Goyal, Itai Gurvich, Jonathan Helm, Ming Hu, Shanshan Hu, Foad Iravani, Srikanth Jagabathula, Ganesh Janakiraman, Fikri Karaesmen, Diwas KC, Saravanan Kesavan, Sang Kim, Song-Hee Kim, Mirko Kremer, Harish Krishnan, Mümin Kurtuluş, Guoming Lai, Cuihong Li, Jun Li, Ilan Lobel, Ruben Lobel, Lauren Lu, Adam Mersereau, Alex Mills, Toni Moreno, Anton Ovchinnikov, Rodney Parker, Ali Parlaktürk, Alfonso Pedraza-Martinez, Ramandeep Randhawa, Paat Rusmevichientong, Soroush Saghafian, Ozge Sahin, Burhaneddin Sandıkçı, Nicos Savva, Melvyn Sim, Amitabh Sinha, Larry Snyder, Greys Sošić, Brad Staats, Robert Swinney, Alireza Tahbaz-Salehi, Gustavo Vulcano, Owen Wu, Wenqiang Xiao, Nan Yang, Zhibin Yang, Fuqiang Zhang, Jiawei Zhang, Yao Zhao, Karen Zheng, and Leon Zhu. The 2014 prize winners are as follows: Two First Prizes Yonatan Gur, Stanford University “Optimization in Online Content Recommendation Services: Beyond Click-Through Rates” Jónas Oddur Jónasson, London Business School “Improving HIV Early Infant Diagnosis Supply Chains in Sub-Saharan Africa: Models and Application to Mozambique” Finalists Antoine Désir, Columbia University “Sparse Process Flexibility Designs: Is Long Chain Really Optimal?” Karthik Murali, University of Illinois at Urbana–Champaign “Municipal Groundwater Management: Optimal Allocation and Control of a Renewable Natural Resource” Yaron Shaposhnik, Massachusetts Institute of Technology “Scheduling with Testing” Linwei Xin, Georgia Institute of Technology “Optimality Gap of Constant-Order Policies Decays Exponentially in the Lead Time for Lost Sales Models”

A M-Learning Content Recommendation Service by Exploiting Mobile Social Interactions

With the rapid development of the Internet and the popularization of mobile devices, participating in a mobile community becomes a part of daily life. This study aims the influence impact of social interactions on mobile learning communities. With m-learning content recommendation services developed from mobile devices and mobile network techniques, learners can generate the learning stickiness by active participation and two-way interaction within a mobile learning community. Individual learning content is able to be recommended according to the behavioral characteristics of the response message of individual learners in the community, and other browsers not of this community are attracted to participate in the learning content with the proposed recommendation service. Finally, as the degree of devotion to the community and learning time increases, the learners’ willingness to continue learning increases. The experiment results and analysis show that individualized learning content recommendation results in better learning effect. In addition, the proposed service proved that the experiment results can be easily extended to handle the recommended learning content for learners’ time-varying interests.

Enhancing User-friendliness of the User Taste Prediction Service Using MapReduce Framework

In our previous work, a user similarity-based contents recommendation service using NFC was proposed for the same goal. This service used a small sample because it used only information of the user who had watched a museum. However, it has been shown that there are some limitations resulting from the difficulty of accurately predicting the user's preference. In order to lift this drawback, this paper introduces a user taste prediction service using big data for improving user-friendliness to a maximum. The proposed service predicts the user's taste using big data such as Twitter and blogs. It is possible to predict the exact user's preference and might recommend more suitable contents to the user's taste because it

Users Viewing Exhibit Similarity-based Contents Recommendation Service by Activity Recognition

We developed a user’s viewing path similarity-based contents recommendation algorithm using LBS (Location Based Services). However, the previous research had a problem that it might recommend contents related with exhibits which users did not like because it divides an exhibition into the areas and recommends contents based on them. To solve this problem, we propose the service based on activity recognition using 3-axis accelerometer sensor. In this service, it is determined by analysis of user’s movements that the corresponding user is moving or viewing. Also, it measures the viewing time that the user spent on viewing exhibits and determines user’s location using LBS. It might recommend more suitable contents to the user's taste because it focuses on the exhibits which users actually watch long.

Hybrid Contents Recommendation Service Using LBS and NFC Tagging

Hybrid Contents Recommendation Service Using LBS and NFC Tagging

Efficient NFC Tagging Pattern-Based Contents Recommendation for Museum Viewers

As various ubiquitous computing technologies become applied into many smart space systems, most museums attempt to use these techniques for their own domain. Near Field Communication(NFC) is one of wireless technologies primarily used for short range communication between smart phones and similar devices, supporting a variety of information exchange services in some effective way. In this paper, we present a new user similarity-based contents recommendation service to improve the counterpart of our previously developed museum viewing system by using NFC. In order to satisfy this goal, this service utilizes similarity of artifact attached NFC tagging patterns of users. Its desirable feature enables users to actually find and obtain their favorite contents by systematically consulting the tastes of the other users very close to those of the first. Also, the proposed system may considerably reduce overloading on contents providing servers by having smart phone clients get certain parts of basic artifact related contents information directly through NFC tags, not the servers.

SNS 기반 전시물 관련 콘텐츠 추천 서비스 설계 및 구현

Abstract As the influence of social networking services across the societies becomes greatly higher, many of the domestic agencies are trying to communicate with users through the introduction of social networking services. In this paper, we present a reliable exhibition-related contents recommendation service to combine social networking service concept with the customized contents recommendation method we previously proposed. The proposed service may effectively and reliably recommend its users exhibition-related contents by exploiting their relationships in the social networks compared with the existing ones. Key Words : Social Network Services, Ubiquitous Computing, Path Searching, Congestion Control, Contents Recommendation * 준회원, 경기대학교 대학원 컴퓨터과학과 ** 정회원, 경기대학교 컴퓨터과학과(교신저자) 접수일자 2012년 1월 12일, 수정완료 2012년 3월 7일게재확정일자 2012년 4월 13일Received: 12 January, 2012; Revised: 7 March 2012Accepted: 13 April 2012 ** Corresponding Author: jhahn@kyonggi.ac.krDept. of Computer Science, Kyonggi University, Korea

Context-Aware Active Services in Ubiquitous Computing Environments

With the advent of ubiquitous computing environments, it has become increasingly important for applications to take full advantage of contextual information, such as the user's location, to offer greater services to the user without any explicit requests. In this paper, we propose context-aware active services based on context-aware middleware for URC systems (CAMUS). The CAMUS is a middleware that provides context-aware applications with a development and execution methodology. Accordingly, the applications based on CAMUS respond in a timely fashion to contextual information. This paper presents the system architecture of CAMUS and illustrates the content recommendation and control service agents with the properties, operations, and tasks for context-aware active services. To evaluate CAMUS, we apply the proposed active services to a TV application domain. We implement and experiment with a TV content recommendation service agent, a control service agent, and TV tasks based on CAMUS. The implemented content recommendation service agent divides the user's preferences into common and specific models to apply other recommendations and applications easily, including the TV content recommendations.