powerpoint presentation on movie recommender system.

movie recommender system

1. MINI PROJECT MOVIE RECOMMENDER SYSTEM PRESENTED BY A.PUTEEN KUMAR(20J21A1201) AMAN PANDEY(20J21A1202) G.NISHANTH REDDY(20J21A1208) JOGINPALLY BR ENGINEERING COLLEGE DEPARTMENT OF INFORMATION TECHNOLOGY UNDER THE GUIDANCE : MR .P.SRINIVAS ASSOCIATE PROFESSIOR

2. contents Abstract Introduction Existing System Proposed Method Module Software Requirement Specification System Design System Life Cycle Result Conclusion

3. ABSTRACT A recommendation system is a system that suggests to users the products. The Movie recommendation system suggests movies a user. By using the dataset and few machine learning algorithms we can build a Recommendation system. It suggests those movies which users liked in the past and would like to watch movies similar to that. The machine learning approach adopted is Content-Based filtering.

4. INTRODUCTION Each and every day millions of users use the internet for various reasons. So, there is a drastic increase in data. When the data is not properly processed, the user when looking for an item has to keep on searching. This provides inefficient results. For this situation, recommendation systems provide relevant recommendations to the user.

5. EXISTING SYSTEM Netflix : Netflix: Netflix uses a recommendation system to suggest movies and TV shows to its users based on their viewing history and preferences. Amazon prime video : Amazon Prime Video also has a recommendation system that suggests movies and TV shows based on the user’s viewing history and ratings.

6. EXISTING SYSTEM IMDb: IMDb offers personalized movie recommendations based on users’ ratings and watch.

7. PROPOSED METHOD Recommendation Algorithm Evaluation User Interface Data collection User Profiling Data Processing

8. MODULES data and recommendation systems providing data to the recommender system approaches for building a recommender system similarity measures

9. DATA AND RECOMMENDATION SYSTEMS a) User Behavior Data : User behavior data is all about the user spending time on a product/ item. This information can be collected from purchase history, ratings and clicks. b) User Demographic Data : User demographic data is all about their information like date of birth, location, age etc. c) Product Attribute Data : Product attribute data contains information about products themselves like cast, crew, genre etc.

10. Providing data to the Recommender System Explicit rating : These are provided by the user when they provide ratings, reviews, likes, feedback. Implicit ratings : Implicit ratings are collected when the user interacts with products. They are inferred from user’s behavior like clicks, purchases, views Product similarity : It is also called item-to-item filtering. It suggests products based on how much the user would like them. User similarity : It is also referred to as user-to-user filtering. This filtering checks the similarity between two users.

11. Approaches for building a Recommender System Popularity based Recommendation system : The products/items are suggested to the user based on popularity. This is not a good way to build a Recommendation system. Content-based Filtering : A content-based filtering suggests user those items/products which are similar to which they have bought in the past. This filtering uses the features of the product to make recommendations.

12. Approaches for building a Recommender System Collaborative-based Filtering : Collaborative filtering recommends users based on the ratings they provided in the past. It collects the ratings of all the users, and the similarity is measured against the users. Hybrid approach : A hybrid approach is a combination of content-based and collaborative-based filtering.

13. Similarity Measures Similarity is measured using the distance between the vectors. Points that are nearest are similar and points which are farthest doesn’t relate to each other. Minkowski distance : We use Minkowski distance when the dimension of the data points are numerical values. 2.Manhattan distance : Manhattan distance calculates the distance between two points measured along its right angle. 3.Euclidean distance : The square root of the sum of squares of the difference between coordinate points which is given by the Pythagorean theorem is how we calculate Euclidean distance

14. Similarity Measures 4. Cosine similarity: It measures the cos angle between two vectors. When cosine with a degree is 0, then its value is 1 so this means the data points are similar. If the cosine degree is 90 the value is 0, so the data points are not similar. 5. Pearson correlation: It measures the correlation between two random variables and it ranges between [-1,1]. If the correlation value is 1 then it is a positive correlation, if the value is -1 then negative correlation.

15. SYSTEM REQUIREMENT SPECIFICATION SOFTWARE REQUIREMENTS: operating system : windows 11 Programming language: Python, Tkinter. Database : PostgreSQL. Machine learning libraries: TensorFlow, PyTorch, and Scikitlearn.

16. SYSTEM REQUIREMENT SPECIFICATION HARDWARE REQUIREMENTS: Processor : Intel i3 RAM : 4 GB Hard disk : 256 GB

17. SYSTEM DESIGN USE CASE DIAGRAM:

18. SYSTEM DESIGN Class Diagram:

19. SYSTEM DESIGN Sequence Diagram:

20. SYSTEM DESIGN Activity Diagram:

21. SYSTEM DESIGN E-R Diagram:

22. SYSTEM LIFE CYCLE Step 1: User has to login/ signup, if the user has already registered then they can simply login using their credential

23. SYSTEM LIFE CYCLE Step 2: When a user is new they have to register and then login. All the details of the new user gets stored in the database table so while they try to login it checks for validity. Step 3: After the user logs in, they would see the following page and request to recommend movies.

24. SYSTEM LIFE CYCLE Step 4: After user successfully logs into the system, they can enter the movie title and ask the system to recommend movies. Step 5: A function reads the movie-list dataset into pandas dataframe. It has the 24 Column. Step 6: As we won’t need all these columns, because some of the data in this dataset is unnecessary, we only need few main features.

25. RESULT

26. RESULT

27. RESULT

28. CONCLUSION Recommendation Systems are widely used in today world for searching for efficient and reliable information. he movie recommendation system in this project recommends movies to the user based on a few features cast, director, genres and keywords.

29. THANK YOU