Weekly Report 5 - Mar 02 2018 - Mar 09 2018

Amlaan

What I did

  • Researched Nested LSTMs for Sentiment Analysis
  • Learned about D3 visualization techniques
  • Experimented with Deck GL for map data visualization

What I plan to do next week

  • More lightweight visualization techniques
  • Integration of weather, Yelp data with other types of data
  • Explore ways to minimize data redundancy and loss in data integration

Blocks or impediments to the plan for next week

  • None for now.

Somshubra

What I did

  • Finish scraping full yelp dataset
  • Finish scraping full Wunderground weather dataset
  • Prepared the codebase for the manual reference datasets of next week

What I plan to do next week

  • See the data stored in the manual datasets.
  • Prepare scripts to clean if required
  • Prepare training of Sentiment models on the reference Yelp dataset when provided.
  • Begin integrating the manual datasets with our collection of data.

Blocks or impediments to the plan for next week

  • Will probably work on this after Tuesday due to midterm.
  • Expect the integration task to go over smoothly. We will assign an integration manager to oversee this task.

Debojit

What I did

  • Setup test project in Django and DRF
  • Got a remote machine on AWS and pushed code to it but without making it public
  • Planned on UI changes

What I plan to do next week

  • Finish server setup and make it testable
  • Finish up backend and test server and move on to help on the front-end.
  • Help finalise what charts are the best for which form of data and test plot it

Blocks or impediments to the plan for next week

  • Setting up some front end design and user flow
  • AWS setup is relatively tedious but nothing unmanageable

Christopher

What I did

  • Explored visualization libraries like D3, bokeh.
  • Explored and worked on understanding yelp data scrapping
  • Attended Project guidance session held by the TAs

What I plan to do next week

  • Work on the next deliverable, manual data set reference.
  • Try out the relevant visualization libraries.
  • Read about sentiment analysis, and ml models.

Blocks or impediments to the plan for next week

  • Identifying the right libraries for each of the task will be an impediment.