Finding local experts from Yelp dataset

Local experts are people who have special expertise in an area, but that expertise is limited to a geographical region. It makes sense to say that someone is a global expert on quantum physics, but it is hard to find someone who knows all about the best ice-cream places anywhere in the world. This is the reason why for local topics, local experts are a better source of information than just experts; they can be a great way to give out a digital word-of-mouth. In this work, we have proposed a system to find local experts on different things from Yelp data. Review and recommendation systems have become a big part of how people consume different products and businesses today. Yelp is a great example of a huge database of reviews on businesses ranging from restaurants, gas stations, salons, and even doctors. The way people consume this extensive database is very much limited to looking at the star rating of the business, which ignores so many other perspectives. We combine various signals from the reviews and spatial data to come up with an algorithm to find the local experts. Yelp provides a large subset of its data for experimentation and we use this dataset to test our hypotheses. Finding local experts can be used in many ways, such as generating weighted, more accurate reviews for businesses, and to create recommendations for new users. If you visit Paris for the first time, you would now be able to get suggestions for food and things to do from native French people who are local experts in those things. The aim of this paper is to automatically find these people who can give you the best local juice on what you want to know.

Name thesis:degree_name

M.S.

Level thesis:degree_level

Thesis

Discipline thesis:degree_discipline

Computer Science

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2015

Author dc:creator

• Jindal, Tanvi

• Local Experts
• Yelp data
• experts

Statement dc:rights

• Copyright 2015 Tanvi Jindal

Language dc:language

en

Handle dc:identifier

http://hdl.handle.net/2142/78499

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/78499