Measuring user’s influence in the Yelp recommender system

Bejarano, Andres; Jindal, Agrima; Bhargava, Bharat

doi:10.1108/prr-02-2017-0016

Cited by 4 publications

(4 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They find that there is a connection between social attributes and user influence. Bejarano et al (2017) also conclude that the findings are relevant in marketing, credibility estimation and Sybil detections, among others.…”

mentioning

confidence: 59%

See 1 more Smart Citation

Editorial

Nurunnabi

2017

PRR

View full text Add to dashboard Cite

The Founding Editor stated that PSU Research Review is seriously committed to maintaining high-quality research (Yamani, 2017). As mentioned in Issue 1, this international journal will enhance global understanding of current research topics among both academics and practitioners (Nurunnabi, 2017). Issue 2 covers computer science and engineering. This issue presents seven papers. Recently, an audacious global cyberattack has targeted vulnerabilities in computer systems in almost 100 countriesone of the largest 'ransomware' attacks involving malicious software (Scott and Wingfield, 2017). In the first paper, Happa and Goldsmith (2017) investigate the description of minor variations in attacks and how and when it may (and may not) be appropriate to communicate those differences in existing attack models. They argue that using annotations appropriately should enable analysts and researchers to express subtle, but important, variations in attacks that may not fit the model of attack that is currently being used. The novelty of the study is in Happa and Goldsmith's demonstration of how annotations may help analysts communicate and ask better questions more rapidly when identifying unknown aspects of attacks (e.g. as a means of storing mental notes in a structured manner, especially while facing zero-day attacks when information is incomplete). The second paper, by Bejarano et al. (2017), argues that 'recommender systems' collect information about users and businesses and describes how this is done in terms of reviews and votes on reviews. They find that there is a connection between social attributes and user influence. Bejarano et al. (2017) also conclude that the findings are relevant in marketing, credibility estimation and Sybil detections, among others. Fang et al. (2017) focus on the problem of generating actionable knowledge from big data. Because existing knowledge bases (KBs) are still far from complete and accurate, they propose a system consisting of two phases, namely, knowledge extraction and truth discovery, to construct a broader KB called 'GrandBase'. They extract new predicates from four types of data sources, namely, Web texts, Document Object Model trees, existing KBs and query streams to augment the ontology of existing KBs (i.e. Freebase). Fang et al. also propose a graph-based approach to conduct better truth discovery for multi-valued predicates. Their study concludes with future research directions regarding GrandBase construction and extension. The fourth paper by Segarra et al. (2017) stresses the importance of internet of things (IoT) applications (e.g. smart homes, smart cities and industry 4.0) for better inventory management using mainly radio frequency identification (RFID) technology. In particular, they focus on RFID technology impairment and the advantages of mature IoT technologies in respect of automatic service discovery, which will be used in our framework to make heterogeneous readers collision free while reading tags. She considers the emerging case of Industry 4.0, where RFID te...

show abstract

mentioning

confidence: 59%

“…The second paper, by Bejarano et al (2017), argues that ‘recommender systems’ collect information about users and businesses and describes how this is done in terms of reviews and votes on reviews. They find that there is a connection between social attributes and user influence.…”

mentioning

confidence: 99%

Editorial

Nurunnabi

2017

PRR

View full text Add to dashboard Cite

show abstract

“…Prior studies show that existing online reviews not only guide people to choose the proper restaurant of their preferences but also determine the tone and flow of upcoming reviews, thus, create homogeneity (Li et al, 2020). So far, hospitality researchers have paid a noticeable amount of attention to Twitter data (Bejarano et al, 2017). The remaining studies that are based upon Yelp reviews mainly analyze mere reviews that (e.g.…”

Section: E-factors Affecting Check-in Behaviormentioning

confidence: 99%

Restaurateur's dilemma: Solved

2022

ASERC

View full text Add to dashboard Cite

The rise of peer-to-peer platforms leads customers to base their opinion on electronic word of mouth more than before, which necessitates the update of factors affecting check-in behavior of customers. Though many industries have been part of this tendency, hospitality industry comes in sight most especially. In this article, the author defines the strongest influences affecting consumer decision-making process when opting for a restaurant. By processing Yelp open-source dataset and applying 9 machine learning models, it becomes clear that Random Forest model is the best tool to predict check-in behavior of diners. On top of that, the number of photos, number of friends are the best electronic factors, while trendiness, kid-friendliness and timing are the most contributing traditional aspects.

show abstract

“…However, the paper is considered to provide a cornerstone of the proposed technique because it found that the two methods can be only applied to select the influencers. Recommender systems of [18] collects information about users and businesses and how they are related. The recommender systems let users share their reviews about products, places, establishments or services.…”

Section: Social Network Aware Recommender Systemmentioning

confidence: 99%

Q2 YouTube: Quantitative and Qualitative Information Analysis based Influencer-aware YouTube Channel Ranking Scheme

Lee¹,

Oh²

2019

Adv. sci. technol. eng. syst. j.

View full text Add to dashboard Cite

With the development of big data, artificial intelligence and deep learning, various social information networks are becoming exponentially intelligent. Of all the various social networks, YouTube is so popular that it is called the YouTube era. Not only video viewers, but also actual video producers, influencer youtubers, are increasing, allowing individuals as well as operators to use and market contents on various topics. As the number of users and contents increases, the choice of information increases, but it is more difficult for individual and business owners to select information that meets their needs. Therefore, this study aims to analyze the channel of the subject that the user needs from various angles and provide the ranking of the channel to individuals and businesses. We crawl the channels and measure the average awareness and influencer youtuber of channels and channel videos by analyzing the qualitative data of quantitative data and comments which are statistical information. As a final case study, we recommend the English learning channel to individual users based on numerical data statistics and emotional analysis results to show the maximum of flipped learning effect regardless of time and space. Plus, we prove the maximum of the effectiveness of marketing by influencer channel which is influential to operators in the medium and long term.

show abstract

Measuring user’s influence in the Yelp recommender system

Cited by 4 publications

References 8 publications

Editorial

Editorial

Restaurateur's dilemma: Solved

Q2 YouTube: Quantitative and Qualitative Information Analysis based Influencer-aware YouTube Channel Ranking Scheme

Contact Info

Product

Resources

About