1 of 15

Restaurant Recommendation System 2nd Semester Project Plan

Parthil Jagani (pjagani2018@my.fit.edu)

Kevin Grondin (kgrondin2018@my.fit.edu)

Jacob Miller (jmiller2018@my.fit.edu)

Isaac Miller (imiller2018@my.fit.edu)

Advisor: Dr. Bhattacharyya (sbhattacharyya@fit.edu)

2 of 15

Team

Parthil Jagani - pjag ani2018@my.fit.edu

Kevin Grondin - kgrondin2018@my.fit.edu

Jacob Miller - jmiller2018@my.fit.edu

Isaac Miller - imiller2018@my.fit.edu

Client: Dr. Bhattacharyya - sbhattacharyya@fit.edu

3 of 15

Goal and Motivation

People have difficulty making decisions as a group

Some voices in groups drown out others, leaving them unheard
Choice paralysis* can occur easily

Our project aims to address these problems in the form of a mobile app

considers the preferences of everyone in a group when deciding where to go for a meal
tries to recommend dining options that best suit the group as a whole

*When we have more choices, we make worse decisions. Having too much choice causes paralysis so we cannot make any decision at all.

4 of 15

Key Features

Users organized within Groups

Groups provide users with a medium to chat with others they plan to meet up with, and allows the system to generate a decision that matches each users’ preferences

Restaurant ranking/preference system

In the context of a decision, a user’s ‘preferences’ are generated procedurally by going through a list of local restaurants, and are then compared with those of other users in the group

Local restaurant data

Restaurant information such as location, images, cuisine type, price range, hours, and more will be accessible to users when looking at specific restaurants

5 of 15

Novel Features

Many of this application’s features have been done before in various contexts

Group messaging
Decision Recommendation Systems

The novelty of our project comes from narrowing the scope and fine-tuning the purpose of these features

Contacts will be primarily local friends and family instead of across the world
Group messages revolve around deciding on and getting together for meals

6 of 15

Technical Challenges

App Development: The project group as a whole has minimal experience with developing programs/applications for mobile devices, so a considerable amount of time will be spent learning app development fundamentals.

Algorithm Evaluation: We will need to thoroughly evaluate decision-making models to find which one has optimal user satisfaction.

Criteria for Decision Making: Deciding which specific categories (i.e. price, distance, cuisine type, etc…) to implement in our decision making algorithm, and how to implement them, will vary depending on the algorithms we test.

7 of 15

System Architecture Diagram

8 of 15

System Architecture Diagram

9 of 15

Evaluation Criteria

For our application, the primary measurement of success/failure will be user feedback

Since our main goal is to provide users with recommendations they will be happy with, the best way to indicate whether or not our app succeeds will be hearing directly from users whether or not they were satisfied.
There are other possible ways to try and measure user satisfaction, but this method is straightforward and tells us exactly what we need to know

While the overall functionality of our app is also important, its success is not entirely tied to how fast it runs or how polished the overall application is, beyond making sure everything works as intended.

10 of 15

Progress Summary

11 of 15

Milestone 4

Gather field research on group restaurant decision-making

Implement, test, and demo different group decision-making algorithms

Continue implementation of group messaging functionality

Create a database of restaurants local to the campus within 10 miles (up to 50 restaurants)

12 of 15

Milestone 5

Continue gathering field research on group decision-making

Evaluate decision-making algorithms on user satisfaction from field research

Implement, test, and demo app feature functionality working together

Create poster for Senior Design Showcase

Create e-book for Senior Design Showcase

13 of 15

Milestone 6

Test and demo entire system

Bug fixing

Final evaluation results

Create user/developer manual

Create demo video