Creating a natural and intutive intrasite Search - BC-331

Preferred Disciplines: Computer Engineering, Data Analytics, Mathematics, Electrical Engineering, Automotive Engineering (Master, PhD, Post-Doc) 
Project length: 8 to 12 months
Desired start date: ASAP
Location: Vancouver, British Columbia; Montreal, Quebec
No. of Positions: 1
Preferences: N/A
Company: PartsAvatar Investments Inc.

About Company:

Parts Avatar is a Canadian startup, geared to revolutionize the Automotive Industry.  Have you ever wondered what a car in the distant future would be like? Cars in the future will be able to drive themselves – and you likely won’t need a license or ownership anymore. They’ll be able to communicate with one another, to help improve road safety. Vehicles will be well integrated with all your devices – like your smartphone, along with live internet connectivity. A touch of a button can send the car away to repair, refill and be ready again at your doorstep! At PartsAvatar, we’re building a foundation for vehicles to be able to communicate, inform and interact like never before.  Our team originally started off with a background in automotive parts and supplies, and has now expanded into providing innovative products and services for the entire automotive industry in Canada. 

Project Description:

An online product is often is referred to with multiple names and spellings – for example, black light, blacklight, UV light, wood’s lamp, all mean the same thing. When a customer types in a misspelling or a variant, they still expect to see a correction with the correct answer. How can we incorporate this intuitiveness in our website search? 

Background and required skills

Research Objectives/Sub-Objectives:

  • The aim of the project is to make the website search intuitive and intelligent
  • Intuitiveness has to be defined in a programatically implementable way – how and when to display corrections and when to go directly to them.à
  • Next, all possible search variations must be mapped exhaustively  - synonyms, mis-spellings, descriptive names, OEM part numbers, related parts, etc.
  • An algorithm should periodically do a search and find these variations on the internet, and then categorize and store them

Methodology:

  • The aim of the project is to make the website search intuitive and intelligent
  • This will directly impact customer experience, customer retention and overall sales
  • First all possible forms of search variations must be itentified  - synonyms, mis-spellings, descriptive names, OEM part numbers, related parts, etc.
  • An algorithm needs to be developed to be able to automatically try and find these variations on the internet
  • These variations must then be fed into the existing part information for indexing, after a manual review process
  • The upgraded search will finally need to be tested and released

Expertise and Skills Needed:

  • Mathematical/Algorithmic/Data analysis skills
  • Strong Java/Javascript coding skills
  • Computer Science/IT Software graduate required

For more info or to apply to this applied research position, please

  1. Check your eligibility and find more information about open projects.
  2. Complete this webform. You will be asked to upload your CV. Remember to indicate the title of the project(s) you are interested in and obtain your professor’s approval to proceed!
  3. Interested students need to get the approval from their supervisor and send their CV along with a link to their supervisor’s university webpage by applying through the webform or directly to Christine Macdonald, cmacdonald(a)mitacs.ca 

Program: