cis3296f24 / Section4-ProjectPresentation

cis3296f24-classroom-project-presentation-ProjectsFeedbackBoard-template created by GitHub Classroom
0 stars 0 forks source link

Streaming Service Referral #10

Open tun24709 opened 2 days ago

tun24709 commented 2 days ago

Project Abstract

This document proposes a recommendation application to allow users to give their opinions on several features of their entertainment purposes, and return a recommended streaming service. The user is able to see checklists for different genres they enjoy, can input titles they strongly favor, as well as titles they have had on their to watch list for some time. In this way, we are able to cross reference several titles from different streaming platforms to see which is most suited to your tastes(as there are a handful to choose from).

Conceptual Design

The design would be similar to a quiz, formatted to asking questions and inputting responses. This would be done from a web browser and would just need a computer. For programming, we would need JavaScript and Python, and use React(could use vue?). and Flask(or Django).

Proof of Concept

https://github.com/annblok/Vue-3-ToDo-List This link does not have the purposes intended but idea of having checked off boxes to communicate interest in different features is main concept (I have drawn out the idea on a Microsoft Whiteboard at the bottom)

API for our purposes: https://api.watchmode.com/docs/#list-titles

Using these two together, we would be able to use one for obtaining user input and use that with the API referenced to provide the best matching streaming service for specified needs. This will vary by person and we would have to figure out best way to use data.

Background

Since Summer 2023, many people have been cancelling Netflix subscriptions due to their limitations on account sharing, causing an influx of people to want to switch to other streaming services. There are many to choose from but often when there are specific shows you want to watch, you'd have to search them up one by one and find where they are available. There is also the issue that people would not know know if the content offered for different genres are well rated such as if Hulu or Netflix had a good selection of horror movies.

Looking this up, there are not really any recommenders but references to how to look at data from different user reviews on tv shows or titles, or just references on the recommendation algorithm Netflix or other streaming services use to help their customers find titles of relevance. The only thing I could really think of is to use data that covers a good amount of streaming services, and we would have to implement our own algorithm of choosing what fits the best for different titles, genres, and more. The API linked above has most functionality with our use as we are able to reference a large amount of data and manipulate it for our uses.

Required Resources

We would need background in Python and JavaScript, we have data from the API but would probably need background on data analysis as to create a proper algorithm that works for our purposes. For using Python it seems using Flask has positive reviews from users for ease of use and Flask frameworks are used for Uber, Netflix, and Reddit which also shows reliability. There is also Django which is also highly rated for the same uses as Flask.

proposal