title
Tutorial 1- Weighted hybrid technique for Recommender system

description
Recommender system becomes very popular and has important role in an information system or webpages nowadays. A recommender system tries to make a prediction of which item a user may like based on his activity on the system. There are some familiar techniques to build a recommender system, such as content-based filtering and collaborative filtering github url: https://github.com/krishnaik06/Recommendation_complete_tutorial Below are the various playlist created on ML,Data Science and Deep Learning. Please subscribe and support the channel. Happy Learning! Deep Learning Playlist: https://www.youtube.com/watch?v=DKSZHN7jftI&list=PLZoTAELRMXVPGU70ZGsckrMdr0FteeRUi Data Science Projects playlist: https://www.youtube.com/watch?v=5Txi0nHIe0o&list=PLZoTAELRMXVNUcr7osiU7CCm8hcaqSzGw NLP playlist: https://www.youtube.com/watch?v=6ZVf1jnEKGI&list=PLZoTAELRMXVMdJ5sqbCK2LiM0HhQVWNzm Statistics Playlist: https://www.youtube.com/watch?v=GGZfVeZs_v4&list=PLZoTAELRMXVMhVyr3Ri9IQ-t5QPBtxzJO Feature Engineering playlist: https://www.youtube.com/watch?v=NgoLMsaZ4HU&list=PLZoTAELRMXVPwYGE2PXD3x0bfKnR0cJjN Computer Vision playlist: https://www.youtube.com/watch?v=mT34_yu5pbg&list=PLZoTAELRMXVOIBRx0andphYJ7iakSg3Lk Data Science Interview Question playlist: https://www.youtube.com/watch?v=820Qr4BH0YM&list=PLZoTAELRMXVPkl7oRvzyNnyj1HS4wt2K- You can buy my book on Finance with Machine Learning and Deep Learning from the below url amazon url: https://www.amazon.in/Hands-Python-Finance-implementing-strategies/dp/1789346371/ref=sr_1_1?keywords=krish+naik&qid=1560943725&s=gateway&sr=8-1 🙏🙏🙏🙏🙏🙏🙏🙏 YOU JUST NEED TO DO 3 THINGS to support my channel LIKE SHARE & SUBSCRIBE TO MY YOUTUBE CHANNEL

detail
{'title': 'Tutorial 1- Weighted hybrid technique for Recommender system', 'heatmap': [{'end': 200.441, 'start': 169.699, 'weight': 0.84}, {'end': 428.028, 'start': 408.954, 'weight': 0.765}, {'end': 528.601, 'start': 467.516, 'weight': 0.746}, {'end': 698.186, 'start': 662.744, 'weight': 0.757}, {'end': 768.475, 'start': 723.642, 'weight': 0.755}, {'end': 981.538, 'start': 836.812, 'weight': 0.74}, {'end': 1265.243, 'start': 1192.823, 'weight': 0.703}], 'summary': 'Tutorial series introduces recommendation systems, covering collaborative, content-based, and algorithm-based systems. it uses datasets from kaggle and discusses creating a movie recommendation engine using the average weighted recommendation based on vote average and vote count, addressing the issue of blank values and multiple features. additionally, it covers analyzing movie data based on weighted average, identifying popular movies, and creating a basic recommendation engine with importance given to popularity and weighted average.', 'chapters': [{'end': 125.534, 'segs': [{'end': 125.534, 'src': 'embed', 'start': 66.121, 'weight': 0, 'content': [{'end': 69.683, 'text': 'K-nearest neighbors and all those different kind of algorithms.', 'start': 66.121, 'duration': 3.562}, {'end': 73.485, 'text': 'But this all will be explained in this particular playlist.', 'start': 70.123, 'duration': 3.362}, {'end': 77.507, 'text': "And this playlist, I'll try to upload daily at least one video on different, different types.", 'start': 73.565, 'duration': 3.942}, {'end': 83.07, 'text': 'Just to begin with, let us just see a basic recommendation system, how we can basically create it.', 'start': 78.127, 'duration': 4.943}, {'end': 88.572, 'text': 'this particular recommendation system is based on average weighted values.', 'start': 83.71, 'duration': 4.862}, {'end': 90.713, 'text': 'now, average weighted values.', 'start': 88.572, 'duration': 2.141}, {'end': 93.234, 'text': "I'll be explaining you what exactly it is.", 'start': 90.713, 'duration': 2.521}, {'end': 100.877, 'text': "the data set that I'm basically considering is basically a movie data set, and this particular movie data set I've downloaded from this Kaggle.", 'start': 93.234, 'duration': 7.643}, {'end': 102.718, 'text': 'it is available freely.', 'start': 100.877, 'duration': 1.841}, {'end': 107.18, 'text': 'what you can do is that you can just copy this particular data set and just provide you the link over here.', 'start': 102.718, 'duration': 4.462}, {'end': 115.787, 'text': "okay, and I'll be uploading this whole Jupyter Notebook file in my GitHub so that you can download and you can basically try it by your own.", 'start': 107.18, 'duration': 8.607}, {'end': 124.313, 'text': "So, initially, to begin with, I'm going to install Pandas and NumPy and then I have, after downloading the dataset from this Kaggle link,", 'start': 116.408, 'duration': 7.905}, {'end': 125.534, 'text': 'I basically have two files.', 'start': 124.313, 'duration': 1.221}], 'summary': 'The playlist will cover different algorithms including a recommendation system based on average weighted values using a movie dataset from kaggle, with the jupyter notebook file available for download on github.', 'duration': 59.413, 'max_score': 66.121, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/_hf_y-_sj5Y/pics/_hf_y-_sj5Y66121.jpg'}], 'start': 1.186, 'title': 'Introduction to recommendation system', 'summary': 'Introduces the new playlist on recommendation system, covering various types including collaborative, content-based, and algorithm-based systems. it aims to upload daily videos on different types, starting with a basic recommendation system based on average weighted values using a movie dataset from kaggle.', 'chapters': [{'end': 125.534, 'start': 1.186, 'title': 'Introduction to recommendation system', 'summary': 'Introduces the new playlist on recommendation system, covering various types including collaborative, content-based, and algorithm-based systems, aiming to upload daily videos on different types, starting with a basic recommendation system based on average weighted values using a movie dataset from kaggle.', 'duration': 124.348, 'highlights': ['The playlist aims to cover different types of recommendation systems including collaborative, content-based, and algorithm-based systems. The chapter mentions the plan to cover various types of recommendation systems, such as collaborative, content-based, and algorithm-based systems.', 'The chapter plans to upload daily videos on different types of recommendation systems. The chapter aims to upload daily videos on different types of recommendation systems to cover a wide range of topics.', 'The initial focus will be on creating a basic recommendation system based on average weighted values using a movie dataset from Kaggle. The chapter explains the initial focus on creating a basic recommendation system based on average weighted values using a movie dataset from Kaggle.', 'The dataset used for the recommendation system is freely available on Kaggle, and the Jupyter Notebook file will be uploaded to GitHub for others to download and try on their own. The chapter mentions the availability of the dataset on Kaggle and plans to upload the Jupyter Notebook file to GitHub for others to download and try.']}], 'duration': 124.348, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/_hf_y-_sj5Y/pics/_hf_y-_sj5Y1186.jpg', 'highlights': ['The playlist aims to cover different types of recommendation systems including collaborative, content-based, and algorithm-based systems.', 'The chapter plans to upload daily videos on different types of recommendation systems.', 'The initial focus will be on creating a basic recommendation system based on average weighted values using a movie dataset from Kaggle.', 'The dataset used for the recommendation system is freely available on Kaggle, and the Jupyter Notebook file will be uploaded to GitHub for others to download and try on their own.']}, {'end': 517.554, 'segs': [{'end': 153.625, 'src': 'embed', 'start': 125.674, 'weight': 0, 'content': [{'end': 136.08, 'text': 'One is pmdb underscore, 5000 credits.csv and then I also have TMDB underscore 5000 movies.csv.', 'start': 125.674, 'duration': 10.406}, {'end': 142.485, 'text': 'So this basically has information about 5000 different movies and they are various,', 'start': 136.12, 'duration': 6.365}, {'end': 147.088, 'text': 'different kind of you know values or features present in this particular data set.', 'start': 142.485, 'duration': 4.603}, {'end': 151.262, 'text': 'like i will just see the data set, what all it is present.', 'start': 147.088, 'duration': 4.174}, {'end': 153.625, 'text': 'so after reading the data set of credits, okay,', 'start': 151.262, 'duration': 2.363}], 'summary': '5000 movies with various features in the dataset.', 'duration': 27.951, 'max_score': 125.674, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/_hf_y-_sj5Y/pics/_hf_y-_sj5Y125674.jpg'}, {'end': 214.78, 'src': 'heatmap', 'start': 169.699, 'weight': 1, 'content': [{'end': 173.179, 'text': 'similarly, I also have true information in this particular data set.', 'start': 169.699, 'duration': 3.48}, {'end': 182.921, 'text': 'I like this particular data set because there is a lot of information in this particular data set when compared to the movie lens data set that we usually use for recommendation system.', 'start': 173.179, 'duration': 9.742}, {'end': 186.642, 'text': "the next thing is that I'm going to also see the head of this movie underscore data frame.", 'start': 182.921, 'duration': 3.721}, {'end': 188.562, 'text': "so for that I'll just go down.", 'start': 186.642, 'duration': 1.92}, {'end': 195.4, 'text': "if I do movies underscore DF dot head, you see that I'm having some information like budget Genres home page.", 'start': 188.562, 'duration': 6.838}, {'end': 198.821, 'text': 'So this is basically the URL of the movie itself.', 'start': 195.78, 'duration': 3.041}, {'end': 200.441, 'text': 'You can go and have a look onto this.', 'start': 198.841, 'duration': 1.6}, {'end': 203.703, 'text': 'There is a unique ID for this particular movies.', 'start': 201.242, 'duration': 2.461}, {'end': 209.285, 'text': 'Keywords, original language, original title, overview.', 'start': 204.823, 'duration': 4.462}, {'end': 212.706, 'text': 'Overview are some features like a brief summary of the movie.', 'start': 209.685, 'duration': 3.021}, {'end': 214.78, 'text': 'So you can see overview.', 'start': 213.78, 'duration': 1}], 'summary': 'The data set contains diverse information, including budgets, genres, and unique ids, useful for recommendation systems.', 'duration': 41.601, 'max_score': 169.699, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/_hf_y-_sj5Y/pics/_hf_y-_sj5Y169699.jpg'}, {'end': 338.106, 'src': 'embed', 'start': 309.441, 'weight': 3, 'content': [{'end': 312.764, 'text': 'sorry, credit data set and 4803 records.', 'start': 309.441, 'duration': 3.323}, {'end': 320.711, 'text': 'similarly, in this, in movies data frame, i have basically 20 um features and four thousand eight hundred and three records.', 'start': 312.764, 'duration': 7.947}, {'end': 325.795, 'text': "now, the first thing that i want to do is that i'll combine this credit and this movies data set.", 'start': 320.711, 'duration': 5.084}, {'end': 327.196, 'text': "there's a reason why i'm doing that.", 'start': 325.795, 'duration': 1.401}, {'end': 328.898, 'text': "i'll just let you know in a while.", 'start': 327.196, 'duration': 1.702}, {'end': 330.099, 'text': 'let me just combine it.', 'start': 328.898, 'duration': 1.201}, {'end': 338.106, 'text': "and to combine it the first of all, the thing that i'll be requiring, that with respect to which column i will try to combine both of this data set.", 'start': 330.099, 'duration': 8.007}], 'summary': 'Combining credit and movies data sets with 4803 records and 20 features.', 'duration': 28.665, 'max_score': 309.441, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/_hf_y-_sj5Y/pics/_hf_y-_sj5Y309441.jpg'}, {'end': 436.754, 'src': 'heatmap', 'start': 408.954, 'weight': 0.765, 'content': [{'end': 414.979, 'text': "on what basis i'm actually merging it on this particular specific id, which is, uh, my feature.", 'start': 408.954, 'duration': 6.025}, {'end': 416.4, 'text': 'right id is basically my feature.', 'start': 414.979, 'duration': 1.421}, {'end': 422.944, 'text': "so i'm merging that and once i merge it you can see the head part of here i'll just have one id column.", 'start': 416.4, 'duration': 6.544}, {'end': 428.028, 'text': "based on that, you'll have you, the whole features will get merged and once you can see that,", 'start': 422.944, 'duration': 5.084}, {'end': 436.754, 'text': 'you can see all the features are merged away and the last, you can basically have some more informations like vote, average, vote out, title cast.', 'start': 428.028, 'duration': 8.726}], 'summary': 'Merging features based on specific id, resulting in unified data with additional information.', 'duration': 27.8, 'max_score': 408.954, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/_hf_y-_sj5Y/pics/_hf_y-_sj5Y408954.jpg'}, {'end': 481.783, 'src': 'embed', 'start': 457.307, 'weight': 4, 'content': [{'end': 463.833, 'text': 'So I just observed some of the uh, you know features like homepage title underscore X, title underscore Y,', 'start': 457.307, 'duration': 6.526}, {'end': 467.376, 'text': 'status production underscore countries which were not that important.', 'start': 463.833, 'duration': 3.543}, {'end': 470.058, 'text': 'So I have basically dropped all this particular columns.', 'start': 467.516, 'duration': 2.542}, {'end': 476.422, 'text': 'Okay And if you just go and see this, uh, this does not make any, Importance with respect to our recommendation.', 'start': 470.258, 'duration': 6.164}, {'end': 481.783, 'text': "So I've just dropped it because there's so many columns over there I want to reduce some of the columns which is not required.", 'start': 476.442, 'duration': 5.341}], 'summary': 'Dropped unnecessary columns to reduce clutter and improve relevance for recommendation.', 'duration': 24.476, 'max_score': 457.307, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/_hf_y-_sj5Y/pics/_hf_y-_sj5Y457307.jpg'}, {'end': 517.554, 'src': 'embed', 'start': 496.906, 'weight': 5, 'content': [{'end': 506.466, 'text': "then I I'm also going to do movies underscore, clean underscore, df.info, and here you can basically see that it is all non null object.", 'start': 496.906, 'duration': 9.56}, {'end': 508.809, 'text': 'that basically means it will not have any tan values.', 'start': 506.466, 'duration': 2.343}, {'end': 514.432, 'text': 'but anyhow, first, always to begin with, go and check whether there is any nan values or not.', 'start': 508.809, 'duration': 5.623}, {'end': 515.092, 'text': 'I verified it.', 'start': 514.432, 'duration': 0.66}, {'end': 517.554, 'text': "there were no nan values, so I'm just going to continue.", 'start': 515.092, 'duration': 2.462}], 'summary': 'Data cleaning process confirmed no nan values present.', 'duration': 20.648, 'max_score': 496.906, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/_hf_y-_sj5Y/pics/_hf_y-_sj5Y496906.jpg'}], 'start': 125.674, 'title': 'Tmdb 5000 movies dataset and movie recommendation engine', 'summary': 'Discusses the tmdb 5000 movies dataset with details on features like cast, budget, revenue, and runtime, comparing it to the movie lens dataset, and introduces the process of creating a movie recommendation engine using the average weighted recommendation based on vote average and vote count, including merging and cleaning the dataset to extract only important features for the recommendation.', 'chapters': [{'end': 238.025, 'start': 125.674, 'title': 'Tmdb 5000 movies dataset', 'summary': 'Discusses the tmdb 5000 movies dataset containing information about 5000 movies, with details on features like cast, budget, revenue, and runtime, comparing it to the movie lens dataset, and highlighting the richness of information present.', 'duration': 112.351, 'highlights': ['The TMDB 5000 movies dataset contains information about 5000 different movies with various features like cast, budget, revenue, and runtime, offering a rich source of data for analysis and modeling.', 'The dataset includes details such as cast in dictionary format, unique IDs for movies, budget, genres, home page URLs, keywords, production companies, release dates, revenue, runtime, spoken languages, and more, providing comprehensive information for analysis and modeling purposes.', 'The chapter compares the richness of information in the TMDB 5000 movies dataset to the movie lens dataset commonly used for recommendation systems, highlighting the substantial amount of information available in the former.']}, {'end': 517.554, 'start': 238.505, 'title': 'Movie recommendation engine', 'summary': 'Introduces the process of creating a movie recommendation engine using the average weighted recommendation based on vote average and vote count, including merging and cleaning the dataset to extract only important features for the recommendation.', 'duration': 279.049, 'highlights': ['Merging the credit and movies dataset based on the movie ID to create a combined dataset with 4803 records and important features for recommendation. Merged the credit and movies dataset based on movie ID, resulting in a combined dataset with 4803 records and essential features for recommendation.', 'Cleaning the dataset by dropping unimportant columns like homepage, title_X, title_Y, status, and production_countries to focus on relevant features for recommendation. Removed unimportant columns like homepage, title_X, title_Y, status, and production_countries to focus on relevant features for recommendation.', 'Ensuring the dataset does not contain any null values, with all features being non-null objects, providing a clean dataset for further processing. Verified that the dataset does not contain any null values, ensuring a clean dataset for further processing.']}], 'duration': 391.88, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/_hf_y-_sj5Y/pics/_hf_y-_sj5Y125674.jpg', 'highlights': ['The TMDB 5000 movies dataset contains information about 5000 different movies with various features like cast, budget, revenue, and runtime, offering a rich source of data for analysis and modeling.', 'The dataset includes details such as cast in dictionary format, unique IDs for movies, budget, genres, home page URLs, keywords, production companies, release dates, revenue, runtime, spoken languages, and more, providing comprehensive information for analysis and modeling purposes.', 'The chapter compares the richness of information in the TMDB 5000 movies dataset to the movie lens dataset commonly used for recommendation systems, highlighting the substantial amount of information available in the former.', 'Merging the credit and movies dataset based on the movie ID to create a combined dataset with 4803 records and important features for recommendation.', 'Cleaning the dataset by dropping unimportant columns like homepage, title_X, title_Y, status, and production_countries to focus on relevant features for recommendation.', 'Ensuring the dataset does not contain any null values, with all features being non-null objects, providing a clean dataset for further processing.']}, {'end': 850.901, 'segs': [{'end': 544.764, 'src': 'embed', 'start': 517.554, 'weight': 1, 'content': [{'end': 520.535, 'text': "but if there, if there is any blank values, I'll try to fix it.", 'start': 517.554, 'duration': 2.981}, {'end': 521.736, 'text': "I'll just show you.", 'start': 520.535, 'duration': 1.201}, {'end': 528.601, 'text': "the today's technique that I'm going to basically use is basically called as weighted average for each movie average rating.", 'start': 521.736, 'duration': 6.865}, {'end': 532.463, 'text': 'now see, guys, there is a feature called as average rating.', 'start': 528.601, 'duration': 3.862}, {'end': 535.145, 'text': 'let me just show it to you.', 'start': 532.463, 'duration': 2.682}, {'end': 538.347, 'text': 'okay, this is the runtime.', 'start': 535.145, 'duration': 3.202}, {'end': 544.764, 'text': 'so many features, so it is becoming very difficult in this particular screen size, you know, to just see.', 'start': 538.94, 'duration': 5.824}], 'summary': "Using weighted average for each movie's average rating to address blank values.", 'duration': 27.21, 'max_score': 517.554, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/_hf_y-_sj5Y/pics/_hf_y-_sj5Y517554.jpg'}, {'end': 590.137, 'src': 'embed', 'start': 558.973, 'weight': 0, 'content': [{'end': 561.356, 'text': 'now, this is the most important thing to understand.', 'start': 558.973, 'duration': 2.383}, {'end': 566.282, 'text': 'uh, this particular formula is basically, uh, taken from this particular source.', 'start': 561.356, 'duration': 4.926}, {'end': 567.964, 'text': 'uh, you know this particular formula.', 'start': 566.282, 'duration': 1.682}, {'end': 577.653, 'text': 'the formula basically says that, uh, w is equal to which is my weighted rating, or r multiplied by v, plus c multiplied by m, divided by v plus m.', 'start': 567.964, 'duration': 9.689}, {'end': 581.434, 'text': 'this is a simple formula of calculating the weighted average.', 'start': 577.653, 'duration': 3.781}, {'end': 586.536, 'text': 'here, r basically means the average for the movie as a number from 0 to 10..', 'start': 581.434, 'duration': 5.102}, {'end': 590.137, 'text': 'if you just go and see the ratings right over here voting average.', 'start': 586.536, 'duration': 3.601}], 'summary': 'The formula for calculating weighted average: w = r*v + c*m / v+m', 'duration': 31.164, 'max_score': 558.973, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/_hf_y-_sj5Y/pics/_hf_y-_sj5Y558973.jpg'}, {'end': 632.753, 'src': 'embed', 'start': 607.466, 'weight': 2, 'content': [{'end': 612.368, 'text': "The average rating is basically the voting average that I've basically considered in this particular example.", 'start': 607.466, 'duration': 4.902}, {'end': 614.668, 'text': 'Okay, so it should be the voting average.', 'start': 612.428, 'duration': 2.24}, {'end': 621.47, 'text': "So what I'm doing is that I'm just using a simple formula saying that average for the movie as a number from 0 to 10.", 'start': 614.788, 'duration': 6.682}, {'end': 624.851, 'text': 'This V, small v basically indicates number of votes for the movie.', 'start': 621.47, 'duration': 3.381}, {'end': 627.952, 'text': 'How many number of votes has been done for that particular movie.', 'start': 624.951, 'duration': 3.001}, {'end': 632.753, 'text': "Votes basically means you're just, you know, how many people have given some ratings for that particular movie.", 'start': 628.112, 'duration': 4.641}], 'summary': 'The average rating is calculated using a simple formula based on the number of votes for the movie, ranging from 0 to 10.', 'duration': 25.287, 'max_score': 607.466, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/_hf_y-_sj5Y/pics/_hf_y-_sj5Y607466.jpg'}, {'end': 686.203, 'src': 'embed', 'start': 662.744, 'weight': 3, 'content': [{'end': 669.367, 'text': 'that does not mean that movie may be wonderful, right, it may be great because only one person has seen that particular movie.', 'start': 662.744, 'duration': 6.623}, {'end': 676.534, 'text': 'so what we do is that we create a m variable which says that minimum votes required to be listed in top 250.', 'start': 669.367, 'duration': 7.167}, {'end': 683.58, 'text': 'so currently, if i consider that there are 3000 movies, at least 250 voting 250 rating should be given.', 'start': 676.534, 'duration': 7.046}, {'end': 686.203, 'text': "then only i'm going to consider those movies right.", 'start': 683.58, 'duration': 2.623}], 'summary': 'At least 250 votes are required for a movie to be listed in the top 250 out of 3000 movies.', 'duration': 23.459, 'max_score': 662.744, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/_hf_y-_sj5Y/pics/_hf_y-_sj5Y662744.jpg'}, {'end': 698.186, 'src': 'heatmap', 'start': 662.744, 'weight': 0.757, 'content': [{'end': 669.367, 'text': 'that does not mean that movie may be wonderful, right, it may be great because only one person has seen that particular movie.', 'start': 662.744, 'duration': 6.623}, {'end': 676.534, 'text': 'so what we do is that we create a m variable which says that minimum votes required to be listed in top 250.', 'start': 669.367, 'duration': 7.167}, {'end': 683.58, 'text': 'so currently, if i consider that there are 3000 movies, at least 250 voting 250 rating should be given.', 'start': 676.534, 'duration': 7.046}, {'end': 686.203, 'text': "then only i'm going to consider those movies right.", 'start': 683.58, 'duration': 2.623}, {'end': 690.787, 'text': 'so similarly, my c variable basically indicates the mean vote across the whole.', 'start': 686.203, 'duration': 4.584}, {'end': 695.503, 'text': "okay, now, what i'll do is that i'll create all these variables.", 'start': 690.787, 'duration': 4.716}, {'end': 698.186, 'text': "first of all, i'll just go ahead and create v.", 'start': 695.503, 'duration': 2.683}], 'summary': 'Minimum 250 votes and mean vote considered for top 250 out of 3000 movies', 'duration': 35.442, 'max_score': 662.744, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/_hf_y-_sj5Y/pics/_hf_y-_sj5Y662744.jpg'}, {'end': 768.475, 'src': 'heatmap', 'start': 723.642, 'weight': 0.755, 'content': [{'end': 726.504, 'text': 'So I can consider movies underscore clean underscore DF.', 'start': 723.642, 'duration': 2.862}, {'end': 728.886, 'text': "I'll take the voting average and I'll just do the mean.", 'start': 726.624, 'duration': 2.262}, {'end': 732.487, 'text': "That way I'll be getting the mean across the whole report.", 'start': 729.406, 'duration': 3.081}, {'end': 735.968, 'text': 'The mean of the movies, mean vote of the movies across the whole report.', 'start': 732.527, 'duration': 3.441}, {'end': 742.491, 'text': 'And then after that I have small m which says that minimum votes required to be listed in the top 250.', 'start': 736.529, 'duration': 5.962}, {'end': 748.933, 'text': "Here, if I want some top 250, here I'm not just going to consider top 250.", 'start': 742.491, 'duration': 6.442}, {'end': 754.455, 'text': "I'll just consider more movies and over here I'm going to use a quantile process which is called a 70 percentile.", 'start': 748.933, 'duration': 5.522}, {'end': 761.29, 'text': "I'll say that only those movies that are more than 70 percentile, right, that value, that count.", 'start': 754.876, 'duration': 6.414}, {'end': 762.551, 'text': "i'm going to just take it up.", 'start': 761.29, 'duration': 1.261}, {'end': 768.475, 'text': 'okay, so till 70th percentile, i mean, uh, it should have more than 70 percentile votes.', 'start': 762.551, 'duration': 5.924}], 'summary': 'Calculating mean vote for movies and considering top 70th percentile votes.', 'duration': 44.833, 'max_score': 723.642, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/_hf_y-_sj5Y/pics/_hf_y-_sj5Y723642.jpg'}, {'end': 855.003, 'src': 'embed', 'start': 827.867, 'weight': 4, 'content': [{'end': 832.608, 'text': 'so this is the most basic recommendation engine i initially came through.', 'start': 827.867, 'duration': 4.741}, {'end': 834.589, 'text': "and, uh, you know, i'm just trying to share your.", 'start': 832.608, 'duration': 1.981}, {'end': 835.769, 'text': 'uh, share you this particular knowledge.', 'start': 834.589, 'duration': 1.18}, {'end': 845.898, 'text': "but as we go ahead with this particular playlist, we'll be developing some complex recommendation system which is basically used in netflix and, uh,", 'start': 836.812, 'duration': 9.086}, {'end': 846.738, 'text': 'many others.', 'start': 845.898, 'duration': 0.84}, {'end': 849.64, 'text': 'uh, you know, recommendation engine itself.', 'start': 846.738, 'duration': 2.902}, {'end': 850.901, 'text': 'so here it is.', 'start': 849.64, 'duration': 1.261}, {'end': 855.003, 'text': "uh, after this, what i'm going to do, i've calculated my weighted average.", 'start': 850.901, 'duration': 4.102}], 'summary': 'Developing a basic recommendation engine, progressing to complex systems used in netflix and others, and calculating a weighted average.', 'duration': 27.136, 'max_score': 827.867, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/_hf_y-_sj5Y/pics/_hf_y-_sj5Y827867.jpg'}], 'start': 517.554, 'title': 'Weighted average movie ratings', 'summary': 'Explores the implementation of the weighted average technique for calculating movie average ratings using the formula w = (r * v + c * m) / (v + m), derived from a particular source, to address the issue of blank values and multiple features. additionally, it discusses the creation of a basic movie recommendation system using a weighted average formula based on voting average and vote count, with a minimum votes requirement of 250, in order to filter out less popular movies and provide a more accurate recommendation.', 'chapters': [{'end': 581.434, 'start': 517.554, 'title': 'Weighted average for movie ratings', 'summary': 'Explores the implementation of the weighted average technique for calculating movie average ratings using the formula w = (r * v + c * m) / (v + m), derived from a particular source, to address the issue of blank values and multiple features.', 'duration': 63.88, 'highlights': ["The technique discussed is the weighted average for each movie's average rating, which addresses the issue of blank values and numerous features.", 'The formula w = (r * v + c * m) / (v + m) is derived from a specific source and is used to calculate the weighted average rating, providing a structured approach to handling movie ratings.']}, {'end': 850.901, 'start': 581.434, 'title': 'Movie rating recommendation system', 'summary': 'Discusses the creation of a basic movie recommendation system using a weighted average formula based on voting average and vote count, with a minimum votes requirement of 250, in order to filter out less popular movies and provide a more accurate recommendation.', 'duration': 269.467, 'highlights': ['Creation of a basic movie recommendation system using a weighted average formula based on voting average and vote count. The speaker discusses the creation of a recommendation system using a weighted average formula based on voting average and vote count, emphasizing the importance of considering both factors in evaluating movie ratings.', 'Minimum votes requirement of 250 to filter out less popular movies and provide a more accurate recommendation. The system sets a minimum votes requirement of 250 to be listed in the top 250, aiming to filter out less popular movies and ensure a more accurate recommendation based on a sufficient number of ratings.', 'Development of complex recommendation systems used in platforms like Netflix. The chapter indicates the potential for future development of complex recommendation systems used in platforms like Netflix, hinting at the progression from the basic system discussed to more advanced algorithms in the field of recommendation engines.']}], 'duration': 333.347, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/_hf_y-_sj5Y/pics/_hf_y-_sj5Y517554.jpg', 'highlights': ['The formula w = (r * v + c * m) / (v + m) is derived from a specific source and is used to calculate the weighted average rating, providing a structured approach to handling movie ratings.', "The technique discussed is the weighted average for each movie's average rating, which addresses the issue of blank values and numerous features.", 'Creation of a basic movie recommendation system using a weighted average formula based on voting average and vote count, emphasizing the importance of considering both factors in evaluating movie ratings.', 'Minimum votes requirement of 250 to filter out less popular movies and provide a more accurate recommendation. The system sets a minimum votes requirement of 250 to be listed in the top 250, aiming to filter out less popular movies and ensure a more accurate recommendation based on a sufficient number of ratings.', 'Development of complex recommendation systems used in platforms like Netflix. The chapter indicates the potential for future development of complex recommendation systems used in platforms like Netflix, hinting at the progression from the basic system discussed to more advanced algorithms in the field of recommendation engines.']}, {'end': 1105.532, 'segs': [{'end': 918.877, 'src': 'embed', 'start': 889.856, 'weight': 0, 'content': [{'end': 891.838, 'text': 'So when I see the top 20 records, here it is.', 'start': 889.856, 'duration': 1.982}, {'end': 894.179, 'text': 'This is the column that I had got.', 'start': 892.518, 'duration': 1.661}, {'end': 902.385, 'text': 'Now, based on the average, this vote count and vote underscore average, along with the moving average weighted average.', 'start': 894.419, 'duration': 7.966}, {'end': 911.411, 'text': 'So, along with this particular terminology that we are using weighted average, we have got our weighted average from a descending order, okay?', 'start': 902.645, 'duration': 8.766}, {'end': 918.877, 'text': 'And this basically recommends that you know suppose you have seen the Shawshank Redemption right?', 'start': 911.872, 'duration': 7.005}], 'summary': 'Analyzing the top 20 records based on vote count, vote average, and weighted average for recommendations.', 'duration': 29.021, 'max_score': 889.856, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/_hf_y-_sj5Y/pics/_hf_y-_sj5Y889856.jpg'}, {'end': 970.852, 'src': 'embed', 'start': 944.524, 'weight': 1, 'content': [{'end': 951.366, 'text': 'so the next movie that can be recommended is basically godfather or fight club, because they are having the weighted average after this one,', 'start': 944.524, 'duration': 6.842}, {'end': 953.605, 'text': 'I mean lesser than 8.3 at least.', 'start': 952.337, 'duration': 1.268}, {'end': 961.327, 'text': 'This was the information about the weighted average and similarly you can plot it.', 'start': 956.144, 'duration': 5.183}, {'end': 966.95, 'text': 'You can plot this particular same weighted average based on ascending is equal to falls.', 'start': 961.727, 'duration': 5.223}, {'end': 970.852, 'text': "You'll be seeing that you'll be having best movies by average goals.", 'start': 967.31, 'duration': 3.542}], 'summary': 'Recommended movies: godfather or fight club, with a weighted average less than 8.3.', 'duration': 26.328, 'max_score': 944.524, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/_hf_y-_sj5Y/pics/_hf_y-_sj5Y944524.jpg'}, {'end': 1051.901, 'src': 'embed', 'start': 1024.926, 'weight': 3, 'content': [{'end': 1030.77, 'text': 'uh, here i was able to see that which is, uh, having the best average rating, and this is based on the number of people voting it.', 'start': 1024.926, 'duration': 5.844}, {'end': 1039.054, 'text': 'right now can i use popularity along with voting count and voting average and i can also perform a recommendation engine right.', 'start': 1030.77, 'duration': 8.284}, {'end': 1039.875, 'text': 'so why?', 'start': 1039.054, 'duration': 0.821}, {'end': 1041.616, 'text': 'why we should use popularity see?', 'start': 1039.875, 'duration': 1.741}, {'end': 1045.738, 'text': 'because there are many popular movies where many people have not been seen right.', 'start': 1041.616, 'duration': 4.122}, {'end': 1048.719, 'text': 'many people have not been given the votings right.', 'start': 1045.738, 'duration': 2.981}, {'end': 1051.901, 'text': 'so for those also, we should not skip the popular movies.', 'start': 1048.719, 'duration': 3.182}], 'summary': 'Identifying best-rated movies by popularity and voting count for better recommendations.', 'duration': 26.975, 'max_score': 1024.926, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/_hf_y-_sj5Y/pics/_hf_y-_sj5Y1024925.jpg'}, {'end': 1097.206, 'src': 'embed', 'start': 1066.067, 'weight': 2, 'content': [{'end': 1068.348, 'text': 'so, based on popularity, you can see that.', 'start': 1066.067, 'duration': 2.281}, {'end': 1074.63, 'text': 'uh, first of all, based on average, uh, votes, you can see that the shawshank redemption was the best movie.', 'start': 1068.348, 'duration': 6.282}, {'end': 1079.232, 'text': 'right, but based on popularity, if you see, minions is the best movie over here.', 'start': 1074.63, 'duration': 4.602}, {'end': 1080.353, 'text': "okay, how i'm plotting it?", 'start': 1079.232, 'duration': 1.121}, {'end': 1081.594, 'text': "i'm basically using seaborn.", 'start': 1080.353, 'duration': 1.241}, {'end': 1086.118, 'text': "first of all, i'm sorting the values based on popularity and i'm plotting it using seaborn.", 'start': 1081.594, 'duration': 4.524}, {'end': 1090.981, 'text': 'okay, so when i see over here, millions is the best popular vote.', 'start': 1086.118, 'duration': 4.863}, {'end': 1097.206, 'text': 'popular movies by vote, sorry, by votes itself, and interstellar, then deadpool, guardian of galaxy.', 'start': 1090.981, 'duration': 6.225}], 'summary': 'The shawshank redemption is the best movie based on average votes, while minions is the most popular based on votes. other popular movies include interstellar, deadpool, and guardians of the galaxy.', 'duration': 31.139, 'max_score': 1066.067, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/_hf_y-_sj5Y/pics/_hf_y-_sj5Y1066067.jpg'}], 'start': 850.901, 'title': 'Movie recommendation analysis', 'summary': "Covers analyzing movie data based on weighted average and using voting count and average to identify popular movies, with 'minions' being the most popular movie based on votes, and a recommendation for the next movie to be recommended based on user behaviors and content-based filtering.", 'chapters': [{'end': 981.538, 'start': 850.901, 'title': 'Analyzing weighted average in movie dataset', 'summary': 'Explains the process of sorting and analyzing movie data based on weighted average, including a recommendation for the next movie to be recommended based on user behaviors and content-based filtering.', 'duration': 130.637, 'highlights': ['The chapter explains the process of sorting the movie dataset based on the weighted average, specifying that the sorting is done in descending order to display the top 20 records with columns like original title, vote count, vote average, weighted average, and popularity.', "It mentions the recommendation of movies like 'Godfather' or 'Fight Club' based on the weighted average and user behaviors, with a specific focus on content-based filtering and the consideration of collaborative filtering based on voting count and voting average."]}, {'end': 1105.532, 'start': 982.338, 'title': 'Movie recommendation analysis', 'summary': "Outlines a data analysis process for movie recommendations, including using voting count and average to identify popular movies and apply a recommendation engine, with 'minions' being the most popular movie based on votes.", 'duration': 123.194, 'highlights': ["Identifying 'Minions' as the most popular movie based on votes The analysis shows that 'Minions' is the best movie based on popularity, as it has received the highest number of votes.", 'Utilizing voting count and average to apply a recommendation engine The chapter discusses using voting count and average to apply a recommendation engine for movie recommendations, enabling the identification of popular movies with high voting counts and averages.', 'Exploring the importance of considering popularity for movie recommendations The chapter emphasizes the significance of not overlooking popular movies, even if they have not been seen or given high ratings, highlighting the need to take popularity into account for comprehensive movie recommendations.']}], 'duration': 254.631, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/_hf_y-_sj5Y/pics/_hf_y-_sj5Y850901.jpg', 'highlights': ['The chapter explains sorting movie dataset based on weighted average in descending order to display top 20 records.', "Recommendation of movies like 'Godfather' or 'Fight Club' based on weighted average and user behaviors.", "Identifying 'Minions' as the most popular movie based on votes.", 'Utilizing voting count and average to apply a recommendation engine for movie recommendations.', 'Emphasizing the significance of not overlooking popular movies for comprehensive movie recommendations.']}, {'end': 1418.839, 'segs': [{'end': 1159.557, 'src': 'embed', 'start': 1127.032, 'weight': 0, 'content': [{'end': 1133.114, 'text': "So I'll just say that recommendation based on scale weighted average and popularity score, the priority is given 50% to both.", 'start': 1127.032, 'duration': 6.082}, {'end': 1139.515, 'text': 'That basically means 50% importance is given to the popularity and 50% importance is given to weighted average.', 'start': 1133.534, 'duration': 5.981}, {'end': 1140.475, 'text': 'And this can be changed.', 'start': 1139.555, 'duration': 0.92}, {'end': 1143.195, 'text': 'You can take 75, 25.', 'start': 1140.535, 'duration': 2.66}, {'end': 1146.296, 'text': "And this is just a basic recommendation engine that I'm creating.", 'start': 1143.196, 'duration': 3.1}, {'end': 1149.677, 'text': 'So it will make sense for you as we go ahead in the upcoming classes.', 'start': 1146.376, 'duration': 3.301}, {'end': 1159.557, 'text': "Now, first of all, before I apply the scale rate average, first of all I'll just do min-max scalar, then transform my weighted average and popularity,", 'start': 1150.848, 'duration': 8.709}], 'summary': 'Recommendation engine prioritizes popularity and weighted average equally at 50% each, but can be adjusted. also includes min-max scalar transformation.', 'duration': 32.525, 'max_score': 1127.032, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/_hf_y-_sj5Y/pics/_hf_y-_sj5Y1127032.jpg'}, {'end': 1267.164, 'src': 'heatmap', 'start': 1190.462, 'weight': 1, 'content': [{'end': 1192.823, 'text': '50 percent importance to the weighted average.', 'start': 1190.462, 'duration': 2.361}, {'end': 1195.925, 'text': "so for that, what i'm doing is that i'll just apply min max scalar.", 'start': 1192.823, 'duration': 3.102}, {'end': 1198.546, 'text': 'i hope everybody knows how you have to apply min max scalar.', 'start': 1195.925, 'duration': 2.621}, {'end': 1202.128, 'text': "so i'm doing fit transform on weighted underscore, average and popularity.", 'start': 1198.546, 'duration': 3.582}, {'end': 1208.951, 'text': "and finally, this is when i normalize df, where i'm combining my scale data frame along with the weighted average and popularity.", 'start': 1202.128, 'duration': 6.823}, {'end': 1211.492, 'text': 'and finally, after scaling it, you will see these are the values.', 'start': 1208.951, 'duration': 2.541}, {'end': 1213.373, 'text': 'the top five records are this.', 'start': 1211.492, 'duration': 1.881}, {'end': 1218.735, 'text': 'now what i will do is that I will take this data set.', 'start': 1213.373, 'duration': 5.362}, {'end': 1220.815, 'text': 'I will convert this into a new data frame.', 'start': 1218.875, 'duration': 1.94}, {'end': 1221.615, 'text': 'Suppose like this.', 'start': 1220.835, 'duration': 0.78}, {'end': 1229.057, 'text': "I'll just create this as a new two features and I'll insert it in our movie clean data set.", 'start': 1221.675, 'duration': 7.382}, {'end': 1230.518, 'text': "So that is what I'm doing over here.", 'start': 1229.217, 'duration': 1.301}, {'end': 1234.119, 'text': "I've created two features, normalized weight average and normalized popularity.", 'start': 1230.538, 'duration': 3.581}, {'end': 1238.32, 'text': 'And I have basically appended movie underscore normalized underscore BX.', 'start': 1234.399, 'duration': 3.921}, {'end': 1244.869, 'text': "Now, once I do this, once I do this, what I have to do is that next step, as I told you, I'll just provide some importance.", 'start': 1239.025, 'duration': 5.844}, {'end': 1246.75, 'text': 'Now see, everybody focus in through this.', 'start': 1245.029, 'duration': 1.721}, {'end': 1250.093, 'text': "I'll declare, I'll create a new feature called a score.", 'start': 1247.251, 'duration': 2.842}, {'end': 1258.939, 'text': 'And in this particular score, I will say that movies underscore clean underscore DF, normalize weight underscore average, and multiply by 0.5.', 'start': 1250.813, 'duration': 8.126}, {'end': 1262.961, 'text': "Now see, I'm multiplying, I'm giving 50% importance to weight average.", 'start': 1258.939, 'duration': 4.022}, {'end': 1265.243, 'text': 'So I basically say multiply by 0.5.', 'start': 1263.001, 'duration': 2.242}, {'end': 1267.164, 'text': 'Similarly, I have to give 0.5 percentage.', 'start': 1265.243, 'duration': 1.921}], 'summary': 'Applying min max scalar, creating new features, and assigning 50% importance to weighted average', 'duration': 76.702, 'max_score': 1190.462, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/_hf_y-_sj5Y/pics/_hf_y-_sj5Y1190462.jpg'}, {'end': 1336.039, 'src': 'embed', 'start': 1308.177, 'weight': 3, 'content': [{'end': 1316.676, 'text': 'if you see now, If you want to get the recommendations first of all, based on the score, Interstellar will be given the recommendation.', 'start': 1308.177, 'duration': 8.499}, {'end': 1318.976, 'text': 'The Minions will be given the recommendation.', 'start': 1317.156, 'duration': 1.82}, {'end': 1321.157, 'text': 'Guardian of the Galaxy will be given the recommendation.', 'start': 1319.036, 'duration': 2.121}, {'end': 1323.477, 'text': 'That is how it is basically done, guys.', 'start': 1321.717, 'duration': 1.76}, {'end': 1327.818, 'text': 'This was two basic examples of this.', 'start': 1324.957, 'duration': 2.861}, {'end': 1336.039, 'text': 'And again, if I try to plot this same thing, after I do this, you can see we have sorted the score.', 'start': 1328.138, 'duration': 7.901}], 'summary': 'Interstellar, minions, and guardian of the galaxy are recommended based on score.', 'duration': 27.862, 'max_score': 1308.177, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/_hf_y-_sj5Y/pics/_hf_y-_sj5Y1308177.jpg'}], 'start': 1106.012, 'title': 'Creating recommendation engines', 'summary': "Discusses creating a basic recommendation engine, emphasizing 50% importance to popularity and weighted average, with values scaled down between 0 to 1. it also covers normalizing the weighted average and popularity, resulting in a new 'score' feature, and explains a recommendation system using weighted average and popularity, with specific importance percentages for columns.", 'chapters': [{'end': 1189.682, 'start': 1106.012, 'title': 'Creating a basic recommendation engine', 'summary': 'Discusses the process of creating a basic recommendation engine, giving 50% importance to popularity and weighted average, which can be adjusted, followed by scaling down the values between 0 to 1 to handle different magnitudes.', 'duration': 83.67, 'highlights': ['The process involves giving 50% importance to both popularity and weighted average, which can be adjusted to different percentages like 75% and 25%.', 'Scaling down the values between 0 to 1 is necessary because popularity and weighted average have different magnitudes, with popularity potentially ranging from 1 to 200 and weighted average ranging from 1 to 10.']}, {'end': 1291.935, 'start': 1190.462, 'title': 'Weighted average and popularity normalization', 'summary': "Discusses the process of normalizing the weighted average and popularity of a dataset, assigning 50% importance to each, resulting in the creation of a new 'score' feature for predicting and analyzing movie data.", 'duration': 101.473, 'highlights': ['The process involves applying min-max scaling to the weighted average and popularity, combining them with the original dataset, and creating new normalized features, resulting in the top five records being displayed with their normalized values.', "Creating a new 'score' feature by assigning 50% importance to the normalized weighted average and 50% importance to the normalized popularity, resulting in a calculated score for each record in the dataset."]}, {'end': 1418.839, 'start': 1292.336, 'title': 'Weighted average recommendation system', 'summary': 'Explains a recommendation system using weighted average and popularity, with interstellar getting the highest importance of 86%, leading to recommendations in decreasing order. it also mentions the possibility of using different importance percentages for columns.', 'duration': 126.503, 'highlights': ["The movie 'Interstellar' is given the highest importance of 86% in the recommendation system, followed by 'Minions' and 'Guardians of the Galaxy', in decreasing order.", "Based on the score, the recommendations are given in the order of 'Interstellar', 'Minions', and 'Guardians of the Galaxy'.", 'The chapter discusses the use of weighted average and the provision of importance based on multiple columns, where 50% importance is assigned to the weighted average and popularity.', 'The possibility of assigning different importance percentages to columns, for example, 25% to one column and 75% to another, is mentioned as a customizable option.']}], 'duration': 312.827, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/_hf_y-_sj5Y/pics/_hf_y-_sj5Y1106012.jpg', 'highlights': ['The process involves giving 50% importance to both popularity and weighted average, which can be adjusted to different percentages like 75% and 25%.', 'The process involves applying min-max scaling to the weighted average and popularity, combining them with the original dataset, and creating new normalized features, resulting in the top five records being displayed with their normalized values.', "Creating a new 'score' feature by assigning 50% importance to the normalized weighted average and 50% importance to the normalized popularity, resulting in a calculated score for each record in the dataset.", "The movie 'Interstellar' is given the highest importance of 86% in the recommendation system, followed by 'Minions' and 'Guardians of the Galaxy', in decreasing order.", "Based on the score, the recommendations are given in the order of 'Interstellar', 'Minions', and 'Guardians of the Galaxy'."]}], 'highlights': ['The process involves giving 50% importance to both popularity and weighted average, which can be adjusted to different percentages like 75% and 25%.', 'The process involves applying min-max scaling to the weighted average and popularity, combining them with the original dataset, and creating new normalized features, resulting in the top five records being displayed with their normalized values.', "Creating a new 'score' feature by assigning 50% importance to the normalized weighted average and 50% importance to the normalized popularity, resulting in a calculated score for each record in the dataset.", "The movie 'Interstellar' is given the highest importance of 86% in the recommendation system, followed by 'Minions' and 'Guardians of the Galaxy', in decreasing order.", "Based on the score, the recommendations are given in the order of 'Interstellar', 'Minions', and 'Guardians of the Galaxy'", 'The chapter explains sorting movie dataset based on weighted average in descending order to display top 20 records.', "Recommendation of movies like 'Godfather' or 'Fight Club' based on weighted average and user behaviors.", "Identifying 'Minions' as the most popular movie based on votes.", 'Utilizing voting count and average to apply a recommendation engine for movie recommendations.', 'Emphasizing the significance of not overlooking popular movies for comprehensive movie recommendations.', 'The formula w = (r * v + c * m) / (v + m) is derived from a specific source and is used to calculate the weighted average rating, providing a structured approach to handling movie ratings.', "The technique discussed is the weighted average for each movie's average rating, which addresses the issue of blank values and numerous features.", 'Creation of a basic movie recommendation system using a weighted average formula based on voting average and vote count, emphasizing the importance of considering both factors in evaluating movie ratings.', 'Minimum votes requirement of 250 to filter out less popular movies and provide a more accurate recommendation. The system sets a minimum votes requirement of 250 to be listed in the top 250, aiming to filter out less popular movies and ensure a more accurate recommendation based on a sufficient number of ratings.', 'Development of complex recommendation systems used in platforms like Netflix. The chapter indicates the potential for future development of complex recommendation systems used in platforms like Netflix, hinting at the progression from the basic system discussed to more advanced algorithms in the field of recommendation engines.', 'The TMDB 5000 movies dataset contains information about 5000 different movies with various features like cast, budget, revenue, and runtime, offering a rich source of data for analysis and modeling.', 'The dataset includes details such as cast in dictionary format, unique IDs for movies, budget, genres, home page URLs, keywords, production companies, release dates, revenue, runtime, spoken languages, and more, providing comprehensive information for analysis and modeling purposes.', 'The chapter compares the richness of information in the TMDB 5000 movies dataset to the movie lens dataset commonly used for recommendation systems, highlighting the substantial amount of information available in the former.', 'Merging the credit and movies dataset based on the movie ID to create a combined dataset with 4803 records and important features for recommendation.', 'Cleaning the dataset by dropping unimportant columns like homepage, title_X, title_Y, status, and production_countries to focus on relevant features for recommendation.', 'Ensuring the dataset does not contain any null values, with all features being non-null objects, providing a clean dataset for further processing.', 'The playlist aims to cover different types of recommendation systems including collaborative, content-based, and algorithm-based systems.', 'The chapter plans to upload daily videos on different types of recommendation systems.', 'The initial focus will be on creating a basic recommendation system based on average weighted values using a movie dataset from Kaggle.', 'The dataset used for the recommendation system is freely available on Kaggle, and the Jupyter Notebook file will be uploaded to GitHub for others to download and try on their own.']}