MOOD MOVIE - A mood based movie recommendation website

Summary

Mah suana introduces a mood-based movie recommendation system designed to enhance user experience by personalizing content according to current emotions. Built with Python and Flask, it uses machine learning to suggest movies that match the user's mood, offering a more tailored experience than traditional streaming services. The project includes stakeholders like end-users, developers, and machine learning engineers, and focuses on user experience, personalization, and content discovery. The system is scalable and secure, with a user-friendly interface and robust backend technologies.

Takeaways

• 😀 The project is a mood-based movie recommendation system aimed at personalizing movie choices according to the user's emotional state.
• 🎬 It addresses the challenge of choosing what to watch from numerous streaming services by suggesting movies that match the user's current mood.
• 🤖 The system uses machine learning algorithms to analyze user mood inputs and recommend movies accordingly.
• 💻 Built using Python and Flask, the project integrates emotional intelligence with traditional recommendation processes.
• 👨‍💻 Stakeholders include end users, software developers, machine learning engineers, project managers, and potential collaborators like streaming services.
• 💡 The project's objectives are to enhance user experience, offer personalization, and facilitate content discovery beyond traditional methods.
• 🛠️ Hardware requirements include a capable GPU, a mid-range processor, and sufficient RAM to support the machine learning model training.
• 🌐 The front-end is developed using HTML, CSS, and JavaScript, focusing on user input, mood processing, and movie recommendation functionalities.
• 🔐 Non-functional requirements such as performance, security, scalability, and usability are emphasized to ensure a robust and user-friendly system.
• 📈 The project timeline includes planning, data collection, model development, validation, and deployment phases, with ongoing monitoring and maintenance.
• 📊 Backend technologies include Python for programming, Pandas for data handling, NumPy for mathematical functions, and Scikit-learn for machine learning model accuracy and validation.

Q & A

What is the main objective of the mood-based movie recommendation system?

The main objective is to enhance the user experience by recommending movies that match the user's current mood, providing a personalized and enjoyable experience compared to traditional methods.

How does the system integrate emotional intelligence into the recommendation process?

The system uses machine learning to understand the user's mood input and recommend movies accordingly, such as suggesting certain movies for a positive, energetic, or reflective mood.

What are the key features of the mood-based movie recommendation system?

Key features include personalization based on mood, content discovery to explore new movies, and a user-friendly interface that allows users to input their mood and receive movie suggestions.

Who are the primary stakeholders of the project?

The primary stakeholders include end users or movie watchers, software developers, machine learning engineers, project managers, and potential collaborators like streaming services.

What are the hardware requirements for the project?

The hardware requirements include a computer with a capable GPU, a mid-range processor with at least four cores, and sufficient RAM, typically 8 GB.

What front-end technologies were used in the development of the website?

The front-end technologies used include HTML for layout, CSS for visual presentation, and JavaScript for interactive elements.

Can you describe the function requirements of the website?

Function requirements include user input for mood, mood processing with filters like mood, language, and genre, and movie recommendation using a trained machine learning model.

What non-functional requirements were considered during the development of the website?

Non-functional requirements include performance, security, scalability, usability, and maintainability to ensure the system is efficient, safe, can handle growth, is user-friendly, and is well-documented for future maintenance.

What programming language and technologies were used for the backend of the project?

Python was the main programming language used, with libraries like Pandas for data handling, NumPy for mathematical functions, and Flask for user interaction in the web framework.

How was the project timeline structured?

The project timeline included planning and preparation, data collection and processing, model development, model validation, and deployment and implementation.

What is the process for a user to get movie recommendations on the website?

Users can log in, select a mood, and use filters like language and genre to get movie recommendations. They can also search for movies by title.

Outlines

00:00

🎥 Introduction to Mood-Based Movie Recommendation System

Mah suana introduces a project that aims to revolutionize movie selection by incorporating emotional intelligence into the recommendation process. The system is designed to understand the user's mood and suggest movies accordingly, using machine learning. Built with Python and Flask, it offers a personalized experience, enhancing user satisfaction and content discovery. The project focuses on improving the user experience, personalization, and content discovery, differing from traditional streaming services that rely on past user history or demographics.

05:01

👥 Stakeholders and Project Overview

The stakeholders of the project include end users who benefit from the personalized movie recommendations, software developers responsible for building the system, machine learning engineers who train the models, project managers overseeing the development process, and potential collaborators like streaming services. The hardware requirements are outlined, including a capable GPU, mid-range processor, and sufficient RAM. The project timeline is detailed, starting from planning and preparation, followed by data collection and processing, model development, validation, and finally deployment and implementation.

10:04

🛠️ Frontend and Backend Technologies

Julia discusses the frontend technologies used in the project, including HTML, CSS, and JavaScript, which provide the layout, visual representation, and interactivity of the website. The backend is primarily developed using Python, with libraries like Pandas for data handling, NumPy for mathematical functions, and Scikit-learn for feature extraction and accuracy validation. Flask is used for user interaction, combining the frontend with the recommendation model. The project timeline includes phases for planning, data collection and processing, model development, and deployment, with an emphasis on the use of tools like the TMDb API for data collection and processing libraries for data manipulation.

🌐 Demo of the Movie Recommendation Website

Julia Pilu provides a demo of the Mood Movie website, showcasing its features and functionality. The website allows users to log in, select a mood, and receive movie recommendations based on their choice. Filters for language and genre are available, and a search bar enables users to find specific movies. The system uses a machine learning model to recommend movies that match the user's mood, providing movie posters, titles, and descriptions. The demo concludes with a logout option, returning to the homepage for new users to sign up or log in.

Keywords

💡Mood-based Recommendation System

A mood-based recommendation system is a type of algorithm that suggests items, such as movies, based on the user's current emotional state. In the video, this system is designed to revolutionize the movie selection process by integrating emotional intelligence. It uses machine learning to match movies to the user's mood, as stated by the speaker: 'it uses machine learning to recommend the movies that match your mood.'

💡Personalization

Personalization refers to tailoring experiences or services to individual preferences or needs. In the context of the video, personalization is a key feature of the mood-based movie recommendation system, aiming to provide a more personalized experience than traditional methods. As Mah suana mentions, 'personalization is the key in this era' and the system 'provides more personalized experience according to your mood.'

💡Machine Learning

Machine learning is a subset of artificial intelligence that enables systems to learn and improve from experience without being explicitly programmed. In the video, machine learning is used to power the recommendation system, allowing it to suggest movies that align with the user's mood. The speaker explains that the system 'uses machine learning to recommend the movies that match your mood.'

💡Content Discovery

Content discovery refers to the process of finding new content that is relevant and interesting to the user. The video highlights how the mood-based recommendation system aids in content discovery by suggesting movies that the user may not have considered before. It is mentioned as one of the main objectives of the project: 'this system helps you to explore or discover new content which you may not have seen before.'

💡Stakeholders

Stakeholders are individuals or groups who have an interest or concern in a project. In the video, the stakeholders of the project include end users, software developers, machine learning engineers, project managers, and streaming services. The speaker outlines the roles and benefits for each stakeholder, such as how end users benefit from the recommendation system and developers ensure technical specifications.

💡Hardware Requirements

Hardware requirements are the physical components needed for a system to function. The video specifies the necessary hardware for the project, including a capable GPU, a mid-range processor with at least four cores, and sufficient RAM, like 8 GB. These are essential for the system's performance, especially for accelerating the learning model training as mentioned by the speaker.

💡Front-end Technologies

Front-end technologies are the tools and frameworks used to create the user interface and user experience of a website. In the video, HTML, CSS, and JavaScript are identified as the main front-end technologies. They are used to structure the layout, style the presentation, and add interactive elements to the website, respectively.

💡User Experience (UX)

User experience (UX) refers to how a person feels when interacting with a system, including ease of use and overall satisfaction. The video emphasizes enhancing user experience as a main objective of the project. The system aims to understand the user's mood to recommend movies that are more likely to be enjoyable, thus improving the user's interaction with the platform.

💡Natural Language Processing (NLP)

Natural language processing is a branch of artificial intelligence that helps computers understand, interpret, and respond to human language in a valuable way. In the video, NLP is used in the recommendation model to process text-based data columns into vectors, which are then used to find movie recommendations that match the user's mood.

💡Flask

Flask is a lightweight web framework for Python that is used to create web applications. In the video, Flask is used to build the user interface of the recommendation system, allowing for user interaction and data handling. The speaker mentions using Flask to combine the front-end and back-end technologies into a functional website.

Highlights

Introduction to a mood-based movie recommendation system that personalizes movie choices according to the user's emotional state.

The system aims to simplify the overwhelming choice of movies on streaming platforms by suggesting content that matches the viewer's mood.

Utilizes machine learning algorithms to analyze user mood and recommend movies accordingly.

Developed using Python and Flask, showcasing a blend of programming and web development technologies.

The system is designed to understand and respond to the user's current mood, enhancing the movie-watching experience.

Aims to provide a more personalized movie-watching experience compared to traditional methods that rely on past history or demographics.

Facilitates content discovery by suggesting new movies that users may not have considered before.

Identifies key stakeholders including end-users, software developers, machine learning engineers, project managers, and potential streaming service collaborators.

Outlines the hardware requirements necessary for the project, such as a capable GPU, mid-range processor, and sufficient RAM.

Front-end technologies used include HTML, CSS, and JavaScript, which contribute to the website's layout, visual representation, and interactivity.

Function requirements of the website include user mood input, mood processing, and movie recommendation generation.

Non-functional requirements focus on performance, security, scalability, usability, and maintainability of the website.

Back-end technologies include Python for programming, Pandas for data handling, NumPy for mathematical functions, and Scikit-learn for machine learning model training.

Project timeline includes planning, data collection, model development, validation, and deployment phases.

Demonstration of the user interface, showcasing the login process, mood selection, and movie recommendation features.

The system allows users to filter recommendations by mood, language, and genre, providing a tailored selection of movies.

Incorporate a search bar for users to look up specific movie titles, enhancing the user experience.

The demo concludes with a logout option, allowing for seamless user account management.

Transcripts

hello everyone I'm Mah suana and today

we'll be talking about a project which

is a mood based M recommendation

system I think everyone likes when

things are personalized according to

their taste right and especially in this

era where personalization is the key our

project aims to revolutionalize the way

we choose movies by integrating

emotional intelligence into the

recommendation process and we always

feel overwhel by all the streaming

services which are provided online right

and you unsure about what to watch so

our project basically solves that you

tell the system your modood and the

system helps you choose according to

your mod the movies it suggest you

certain movies which will be according

to the mo mod you've given input let's

say positive energetic or reflexive it

uses machine learning to recommend the

movies that match your mood it is built

using Python and flas so our Bo based

movie recommendation system it is a

cutting it solution which is designed to

understand and respond to the users's

current mood providing a tailored movie

suggestion that can enhance the user

experience in contrast with the

traditional streaming services which are

available online that uses uh the users

past history or the demographies our

project are for the purpose of our mood

based recommendation system is to

develop a system that will suggest the

content based on the user's current mood

so it uh these are the main objective

which our project focuses on the first

thing is the to enhance the user

experience so basically we are

understanding the user's mood the system

can recommend movies and content which

is more likely to be enjoyable to the

user the second thing is personalization

as I said personalization is the key in

this era so mood based movie

recommendations it provides more more

personalized experience according to

your mood than the traditional methods

and next is the content Discovery this

system helps you to explore uh or

discover new uh content which you may

not have seen before okay now so let's

talk about the stakeholders of a project

the first ones are the end users or the

movie Watchers they are directly

benefited from the recommendation system

and their satisfaction with the user

interface next ones are the software

developers the builders they basically

translate the project requirements into

a functional system it ensures technical

specifications and adares the best

practices and it is scalable for the

future grou the next ones are the

machine learning uh Engineers they train

and maintain the machine learning models

they provide accurate and more

personalized

recommendations next the next one the

next stakeholders are the project

managers project managers are basically

the coordinators which oversee the

entire development process of a project

let's say the time people required or

the budget required Etc next are the uh

streaming services the streaming

services we can say they can be a

potential collaborator in future if the

system integrates with an existing

platform then it can take it to the

different level let us look into the

hardware requirements of a project the

first we need a computer which has a

capable GPU next we'll require a

processor which is a mid-range processor

with at least four CS the next we need a

sufficient Ram which is available in our

laptops or PCS let's say 8 GB and the

and then we'll require a high

performance Hardware to accelerate the

learning model training hello everyone

I'm Julia and I'm going to explain the

front Technologies we've used in our

website so for our front end the three

main things and elements that you used

for our website include HTML CSS and

JavaScript hm has basically given us a

layout and a framework for our website

and we used different elements in it

like header main uh elements button Etc

to create an overall layout for a webite

the next element that we've used is CSS

which defines the visual representation

and presentation of our web page and

controlling the layout styling and

animation it creates a visually

appealing overall look of the website so

that it gives a modern look and touch to

our website so these are the function

requirements of our website which

include user input mode processing and

movie recommendation the first and

foremost function requirement is user

input where the user enters their mode

and we generate a movie based on their

inputs mode processing so in our website

we have different filters which

basically include a mode filter a

language filter and a genre filter where

you can select as for your requirements

and we will generate a movie based on

those requirements and we also have an

additional search bar in case you want

to search for a movie of you're liking

or choosing and the last part is movie

recommendation where we recommend you a

movie that's suitable for your mood and

customized for you using our train

trained machine learning model to

generate a movie recommendation so these

is the nonfunctional requirements of our

website which include performance

security scalability usability and

manageability performance ensures that

our system can generate good movie

recommendations quickly efficiently and

accurately and we've also implemented

robust security measures so that user

data is safe and protected our design is

also scalable so that it can handle a

growing number of users as well as

images and our website is also intuitive

and user friendly so that users can use

it with ease and maintainability we've

developed our website by with good

proper documentation so it can be

maintained and kept clean for future use

as well hello everyone this is ID and

I'll be explaining the backend

technologies that we use for

project okay so our main programming

language for this project was python we

chose python because it was like a on

stop for all our requirements and was

the most appropriate language where we

could use all our other Technologies and

models for the recommendation system and

in our recommendation model we used

fandal for data handling and then we use

num for mathematical functions in our

model recommendation we use escalar

feature extraction and metrics for

accuracy and validation and in flask we

did our user indication so using flask

we could combine our and fr to have a

function website so I'll be giving you

all a brief project timeline that we use

so our first phase is planning and

preparation and the second phase was

data collection and processing which was

done by our teammate M the third one was

model development and followed by

valuation validation of the model the

last one was deployment and

implementation of the model let me

explain in detail the first phase

planning and ption so in this phase we

basically scope out the game and

objective of our project and then we

decide who does what and we decide

theate resources needed for this project

the Second Step data collection and pre

processing so in this we used our tmb

website from kagle and we had to clean

the data and reprocess it to fit our

model followed by model development we

used skarn feature extraction and a sub

field of machine learning that is

natural language processing the model

was deployed after that using pickle

module the validation and validation we

use a scal metric accuracy and

validation methods and then the last one

is monitoring and maintenance in this

basically what we do is we update our

database and our software based on how

and how the user so let's talk a little

bit more about our development tools we

use development tools mainly for two two

phases that is our data collection and

processing and recommendation algorithms

so I like I just said our database is

from the tmtv website and we weped to

get database we use the tmtv API to get

our post links for the website and then

we use data processing librar p and

which are B Library which are used for

basically manipulating the data as for

the recommendation algorithms we use

natural language processing under

natural language processing we use the

tfid vectorizer we convert our text

based data colums into vectors using the

cosine similarity we then associate the

movie the value and the closest Value

and movies having Val similar to that

are to

the hello everyone my name is Julia pilu

and I will be showing you guys a demo on

our project the Moon movie website let

me Begin by presenting my

screen okay so this is our landing page

or our homepage

and this is our footer section along

with our service

cards so let's get

started we get started by clicking on

that

button and if you are an existing user

you can log in and if you are a new user

you'll have to sign up since I already

have an account made let me log in

and let me enter my

password

login so this is my user

profile where I can get my movie

recommendations generated based on my

mode so this is the core functionality

of our website and it has filters like

movie name JRA language and mode let me

Begin by selecting a

mode let's say relax

in and I click on the recommend button

and it gives me a list of movie

recommendations for my mood that's

relaxing and what it basically does is

it gives you a movie poster along with

the movie title and a short description

about the

movie let's check out some other

filters so let's say I want to watch a

Hindi movie

I click on the Hindi language filter and

let's suppose my mode is

neutral and let me

recommend so these are the movies that

it recommends for me where I can watch

them whenever I want in whatever mode

they're all classic

movies so let me go

back and you don't have to select both

filters at the same time you can only

select one filter for example over here

I've selected only the language filter

and I've made it so that the language is

Spanish so it recommends a list of

Spanish movies for

me and I can pick whichever one I want

among these and watch

them I can also select based on um genre

so let's say comedy

and it recommends a mixture of all

comedy movies for me for a variety of

languages it's not just limited to

English

movies and suppose I want to look up a

movie based on its title let's say

rashar and it gives you a list of all

the rashar movies in sequence it gives

you the movies title and the

description in case I want to know which

rashar movie to watch

again so yeah this is our movie

recommendation machine learning

model and now I can go back and log

out and it takes me back to our homepage

where another user can sign up login or

I can repeat the same process

again so this was it for the demo thank

you guys so much for listening to me