Download as PPTX, PDF, TXT or read online from Scribd
Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1/ 19
Movie
RecommendationSyste m
2003209 - Raghav Rathi
2003210 - Meer Kukreja 2003211 - Aniket Lilwani Mentor - Mrs. Juhi Janjua Introduction Recommendation System produces a ranked list of items on which a user might be intrested, in the context of his/her current choice of an item.A recommendation system is the information filtering system. The function of recommeder system is to provide us the best matching information relevant to our requirement from a bigger set of information provided.Our Recommendation system makesI use of the movies watched by a user and recommends a movie by taking into account of the overview,crew,cast,etc. Our system uses the concept of Collborative filtering to generate the desired output.
Recommender Systems has mainly two parts Item and User.
• The most common types of recommendation systems are content-based and collaborative filtering recommender systems. • In collaborative filtering, the behavior of a group of users is used to make recommendations to other users. The recommendation is based on the preference of other users. A simple example would be recommending a movie to a user based on the fact that their friend liked the movie. • There are two types of collaborative models Memory-based methods and Model-based methods. The advantage of memory- based techniques is that they are simple to implement and the resulting recommendations are often easy to explain. They are divided into two:
• ·User-based collaborative filtering: In this model, products are recommended to a user based on the fact that the products have been liked by users similar to the user. For example, if Derrick and Dennis like the same movies and a new movie come out that Derick like, then we can recommend that movie to Dennis because Derrick and Dennis seem to like the same movies. • Item-based collaborative filtering: These systems identify similar items based on users’ previous ratings. For example, if users A, B, and C gave a 5-star rating to books X and Y then when a user D buys book Y they also get a recommendation to purchase book X because the system identifies book X and Y as similar based on the ratings of users A, B, and C. Recommendation Systems
Collaborative Hybrid (CF +
Content - Based Filtering Content Based Approach (CF) Arppaoch) Objectiv • The main objective of our recommendation systems is to provide recommendations
e based on viewed information of the users' preferences. • This systems makes uses of information filtering techniques to process information and provide the user with potentially more relevant items. • The main aim of our project is to provide a highly accurate recommendation with a clean and a decent UI design making it feasible for any user to use without any hassles. • Our goal is to find a new way to understand how classification works and improve the requirement of improving content-based recommender systems. Literature Review • Recommendation System is a vast area which is used everywhere in every field. • People use recommendations as it saves time, so it plays a vital role in various areas. It is used in many real life applications like Entertainment, E-Commerce, Services, Social Media etc. • Main goal of recommendation system is to forecast the rating which a specific user gives to an item. • It helps the user to find the optimum recommendation from the available list of items. In order to achieve the goal of the project, • The first process is to do enough background study, so the literature study was conducted. • The whole project is based on a big amount of movie data so that we choose quantitative research.Researching and choosing data which is highly relevant to our model is the utmost important thing and should be done carefully. • In Some previous Existing System due to less validation of data Null Values or irrelevant values appear in major case which fails the system. • Many existing movie recommendation systems are developed based on different algorithms. Thus, it is crucial to understand the state-of-the-art developments of these systems, their advantages and disadvantages as well as areas which require more research. • We conducted a literature review on the existing solutions for movie recommender systems, describe and compare them based on various features, and present future research directions.This gave us a breif idea of how to start Problem Statement • To build an effective recommender system is also a challenge because user’s preference keeps on changing with time it's uncertain . • Proposed system should be have a constant accuracy and Minimal false suggestions. • To have a clean UI which makes it easy for the user to get used to it Block Diagram Algorithm for Proposed System Step 1 : Search for Dataset relevant to your choice of System. Step 2: Remove Unwanted and Nullified Values from the dataset for better results. Step 3: Choose the Appropriate tags highly relevant to your system. Step 4: Create a concatenated String of Tags & convert them all to lower Step 5: Perform text vectorization over that Concatenated Strings Step 6: Evaluate cosine similarity and generate a Similarity Matrix and sort it in decreasing order of the similarity Step 7 : Extract the top 5 movies relevant to the selected movie and provide it to the pickle object Step 8 : Search for the movie and display top 5 relevant movies using Streamlit library Flow Chart Text Vectorization: Bag of Words
• The process of converting words into numbers are called Vectorization.
• Julie loves John more than Linda loves John Item 1: [2, 0, 1, 1, 0, 2, 1, • Jane loves John more than Julie loves John Item 2: [2, 1, 1]1, 0, 1, 1, 1, • The cosine angle (the smaller the angle) between the two vectors'1] value is 0.822 which is nearest to 1. • Bag of words is a text vectorization technique that converts the text into finite length vectors. The boW model is easy to implement and understand. Bag of words has few drawbacks, which can be overcome by using advanced techniques. • Cosine Similarity measures the cosine of the angle between two non-zero vectors of an inner product space. • This similarity measurement is particularly concerned with orientation, rather than magnitude. • Two cosine vectors that are aligned in the same orientation will have a similarity measurement of 1, whereas two vectors aligned perpendicularly will have a similarity of 0. • If two vectors are diametrically opposed, meaning they are oriented in exactly opposite directions (i.e. back-to-back), then the similarity measurement is -1. • Often, however, Cosine Similarity is used in positive space, between the Screenshots References • https://towardsdatascience.com/cosine-similarity-how-does-it-measure-the-similarity-maths- behind-and-usage-in-python-50ad30aad7db
