Recommender Systems for Personalized User Experience: Lessons learned at Booking.com

2021 ◽  
Author(s):  
Ioannis Kangas ◽  
Maud Schwoerer ◽  
Lucas J Bernardi
2010 ◽  
pp. 1-22
Author(s):  
Nalin Sharda

Modern information and communication technology (ICT) systems can help us in building travel recommender systems and virtual tourism communities. Tourism ICT systems have come a long way from the early airline ticket booking systems. Travel recommender systems have emerged in recent years, facilitating the task of destination selection as well activities at the destination. A move from purely text-based recommender systems to visual recommender systems is being proposed, which can be facilitated by the use of the Web 2.0 technologies to create virtual travel communities. Delivering a good user experience is important to make these technologies widely accepted and used. This chapter presents an overview of the historical perspective of tourism ICT systems and their current state of development vis-à-vis travel recommender systems and tourism communities. User experience is an important aspect of any ICT system. How to define user experience and measure it through usability testing is also presented.


2017 ◽  
Vol 35 (1) ◽  
pp. 120-143 ◽  
Author(s):  
Judith Wusteman

Purpose The purpose of this paper is to describe the process and implications of usability testing a prototype version of the Letters of 1916 Digital Edition. Design/methodology/approach The paper presents the testing, the lessons learned and how those lessons informed the subsequent redesign of the site. Findings Results imply that a majority of users, even digital humanists, were not looking for a unique and specialised interface, but assumed – and preferred – a user experience that reflects common search systems. Although the audience for digital humanities sites is becoming increasingly diverse, the needs of the different user groups may be more similar than had previously been assumed. Research limitations/implications The usability test employed 11 participants, five of whom were coded as “general public”. Four of these five had previously volunteered to transcribe and upload letters. This meant that they were already familiar with the project and with the Letters of 1916 Transcription Desk. However, their prior involvement was a result of their genuine interest in the site, thus ensuring that their interactions during testing were more realistic. Practical implications The lesson learned may be useful for the Digital Editions of future crowdsourced humanities projects. Originality/value Letters of 1916 is the first crowdsourced humanities project in Ireland. The theme of the project is topical, emotive and socially important in Ireland and among Irish diaspora today. The project’s content has been created by the “ordinary citizens of Ireland” and they are likely to be the major users of the Digital Edition. The study explores how the Digital Edition can support these users, while also facilitating the range of traditional scholars and digital humanities researchers.


AI Magazine ◽  
2022 ◽  
Vol 42 (3) ◽  
pp. 7-18
Author(s):  
Harald Steck ◽  
Linas Baltrunas ◽  
Ehtsham Elahi ◽  
Dawen Liang ◽  
Yves Raimond ◽  
...  

Deep learning has profoundly impacted many areas of machine learning. However, it took a while for its impact to be felt in the field of recommender systems. In this article, we outline some of the challenges encountered and lessons learned in using deep learning for recommender systems at Netflix. We first provide an overview of the various recommendation tasks on the Netflix service. We found that different model architectures excel at different tasks. Even though many deep-learning models can be understood as extensions of existing (simple) recommendation algorithms, we initially did not observe significant improvements in performance over well-tuned non-deep-learning approaches. Only when we added numerous features of heterogeneous types to the input data, deep-learning models did start to shine in our setting. We also observed that deep-learning methods can exacerbate the problem of offline–online metric (mis-)alignment. After addressing these challenges, deep learning has ultimately resulted in large improvements to our recommendations as measured by both offline and online metrics. On the practical side, integrating deep-learning toolboxes in our system has made it faster and easier to implement and experiment with both deep-learning and non-deep-learning approaches for various recommendation tasks. We conclude this article by summarizing our take-aways that may generalize to other applications beyond Netflix.


2021 ◽  
Author(s):  
Emily Perchlik ◽  
Donald MacDonald

<p>North American bridge design is dominated by a culture of risk aversion and economic constraint. While objectives of safety and efficiency should be the baseline of any project, they are sometimes set as the sole benchmarks for a successful bridge design within the North American context. When the end game is to simply meet the baseline of safety and efficiency, goals related to user experience and aesthetic impacts are often considered superfluous. This paper showcases lessons learned from designing within this context.</p><p>Stories from bridge designs showcase the ups and downs of bootstrapping higher design goals into footbridge projects in the Wild West.</p>


Information ◽  
2019 ◽  
Vol 10 (5) ◽  
pp. 174 ◽  
Author(s):  
Diego Monti ◽  
Enrico Palumbo ◽  
Giuseppe Rizzo ◽  
Maurizio Morisio

Recommender systems have gained a lot of popularity due to their large adoption in various industries such as entertainment and tourism. Numerous research efforts have focused on formulating and advancing state-of-the-art of systems that recommend the right set of items to the right person. However, these recommender systems are hard to compare since the published evaluation results are computed on diverse datasets and obtained using different methodologies. In this paper, we researched and prototyped an offline evaluation framework called Sequeval that is designed to evaluate recommender systems capable of suggesting sequences of items. We provide a mathematical definition of such sequence-based recommenders, a methodology for performing their evaluation, and the implementation details of eight metrics. We report the lessons learned using this framework for assessing the performance of four baselines and two recommender systems based on Conditional Random Fields (CRF) and Recurrent Neural Networks (RNN), considering two different datasets. Sequeval is publicly available and it aims to become a focal point for researchers and practitioners when experimenting with sequence-based recommender systems, providing comparable and objective evaluation results.


Author(s):  
Wen-Yau Liang ◽  
Chun-Che Huang ◽  
Tzu-Liang Tseng ◽  
Zih-Yan Wang ◽  
◽  
...  

Introduction. Measuring user experience, though natural in a business environment, is often challenging for recommender systems research. How recommender systems can substantially improve consumers’ decision making is well understood; but the influence of specific design attributes of the recommender system interface on decision making and other outcome measures is far less understood. Method. This study provides the first empirical test of post-acceptance model adaption for information system continuance in the context of recommender systems. Based on the proposed model, two presentation types (with or without using tag cloud) are compared. An experimental design is used and a questionnaire is developed to analyse the data. Analysis. Data were analysed using SPSS and SmartPLS (partial least squares path modeling method). Statistical methods used for the questionnaire on user satisfaction were a reliability analysis, a validity analysis and T-tests. Results. The results demonstrate that the proposed model is supported and that the visual recommender system can indeed significantly enhance user satisfaction and continuance intention. Conclusions. In order to improve the satisfaction or continuance intention of users, it is required to improve the perceived usefulness, effectiveness and visual attractiveness of a recommender system.


Author(s):  
Juliana Alvarez ◽  
David Brieugne ◽  
Pierre-Majorique Léger ◽  
Sylvain Sénécal ◽  
Marc Frédette

Recent research has called for the use of enriched measures, that is, psychophysiological measures of emotional and cognitive states, in user experience (UX) testing. This chapter investigates how these enriched measures can inform user experience evaluation while maintaining agility and speed in managing UX projects. Using a multiple case approach, this chapter presents the analysis of 12 recent user experience projects in which enriched measures were used. Lessons learned with regard to challenges encountered are outlined. They emphasize on: (1) the nature of the research question impacts the completion time and the complexity of the project; (2) the need to communicate and coordinate between all parties; (3) the need to anticipate the collected measurements and enhanced results using a mosaic of hybrid collection methods; (4) the nature of the results adapted to underline the operational side without reducing the quality of the work performed; and (5) the time constrains influenced and influencing the pre-tests and project’s granularity. This chapter concludes with lessons learned from an agile/UX development approach in the realization of Sprint projects.


Sign in / Sign up

Export Citation Format

Share Document