1 00:00:01,290 --> 00:00:02,340 Hey, welcome back. 2 00:00:02,340 --> 00:00:10,350 In this video, we will explore the data sets we have so you can find these three attached in the lecture 3 00:00:10,350 --> 00:00:10,900 resources. 4 00:00:10,900 --> 00:00:16,050 So you have a credit dot csv, file, movies, dot CSV and ratings dot CSV. 5 00:00:16,320 --> 00:00:22,230 Let's open first movies dot csv to see what this is. 6 00:00:22,530 --> 00:00:24,930 So we have multiple columns here. 7 00:00:24,930 --> 00:00:29,070 You can open this with Excel or any other spreadsheet program. 8 00:00:29,640 --> 00:00:31,770 So we have a budget column. 9 00:00:31,770 --> 00:00:35,370 This is obviously the budget of the movie in dollars. 10 00:00:35,400 --> 00:00:40,050 The genre genre is a list of dictionaries. 11 00:00:41,040 --> 00:00:50,250 So because a movie can have multiple genres, it is a list with the ID of the genre and the name, for 12 00:00:50,250 --> 00:00:51,420 example, action. 13 00:00:51,420 --> 00:00:56,160 And then we have another dictionary with the ID of the adventure genre. 14 00:00:56,160 --> 00:01:01,090 So it's a list of genres, basically fantasy science fiction. 15 00:01:01,090 --> 00:01:05,440 So that's for the movie with this home page. 16 00:01:05,440 --> 00:01:14,080 So we also have this home page column as well, the ID of the movie and some keywords. 17 00:01:14,170 --> 00:01:17,830 So the keywords is also a list of dictionaries. 18 00:01:18,320 --> 00:01:23,000 And keywords also have an ID and of course the name. 19 00:01:23,010 --> 00:01:26,210 So culture clash, for example, future. 20 00:01:26,210 --> 00:01:30,440 And so this tag may be repeated in other movies as well. 21 00:01:30,440 --> 00:01:35,600 So basically the future tag will have the same ID in all other movies. 22 00:01:36,500 --> 00:01:40,310 We may need these data when we do. 23 00:01:40,340 --> 00:01:47,630 Content based filtering, for example, when we want to compare the tags of different movies to find 24 00:01:47,660 --> 00:01:54,050 similar movies to recommend to a user who clicked on a particular movie. 25 00:01:54,980 --> 00:01:57,320 So for content based filtering. 26 00:01:58,220 --> 00:02:01,760 We have the language of the movie and then the title of the movie. 27 00:02:01,760 --> 00:02:03,620 So the first one is Avatar. 28 00:02:04,400 --> 00:02:10,669 The overview field will also be very useful when we do content based filtering. 29 00:02:10,850 --> 00:02:17,720 So it gives you the description of the movie, the popularity score. 30 00:02:17,750 --> 00:02:20,330 So how popular this movie is. 31 00:02:23,620 --> 00:02:26,800 Then we have the list of production companies. 32 00:02:26,830 --> 00:02:28,720 Also a list of dictionary. 33 00:02:31,620 --> 00:02:33,370 Production countries as well. 34 00:02:33,390 --> 00:02:34,980 A list of dictionaries. 35 00:02:35,370 --> 00:02:38,910 Release date of the movie, the Revenue. 36 00:02:38,910 --> 00:02:40,770 So how much money it made. 37 00:02:40,890 --> 00:02:42,180 Runtime. 38 00:02:42,390 --> 00:02:44,700 That's the time in minutes. 39 00:02:46,590 --> 00:02:50,180 Spoken languages in the movie inside the movie. 40 00:02:50,190 --> 00:02:52,710 So there may be more than one language. 41 00:02:52,710 --> 00:02:55,020 The status release tagline. 42 00:02:55,920 --> 00:02:57,070 Title again. 43 00:02:57,090 --> 00:02:58,380 It's repeated. 44 00:02:58,410 --> 00:03:02,910 Anyway, that's something you can clean up with pandas. 45 00:03:03,630 --> 00:03:05,610 The average number of votes. 46 00:03:05,610 --> 00:03:07,800 So that's from 0 to 10. 47 00:03:08,100 --> 00:03:09,410 And then the vote count. 48 00:03:09,420 --> 00:03:10,530 So how many votes? 49 00:03:10,530 --> 00:03:15,270 And the average score that was movies dot CSV. 50 00:03:15,270 --> 00:03:20,460 And then we have credit dot CSV here. 51 00:03:20,460 --> 00:03:24,390 We will find some additional data about the movies. 52 00:03:24,570 --> 00:03:31,620 So it's the same database of movies and the two databases share the same ID column. 53 00:03:31,620 --> 00:03:34,200 So that's the ID of the movie Avatar. 54 00:03:34,320 --> 00:03:36,450 And also the title column is the same. 55 00:03:36,450 --> 00:03:40,170 But then we have some additional information here, such as the cast. 56 00:03:40,170 --> 00:03:48,270 So the names of the actors, the crew, which helped to produce the movie. 57 00:03:48,570 --> 00:03:56,500 So these are some additional data and we may need to join to merge these two files into one single dataframe 58 00:03:56,500 --> 00:03:57,340 in Python. 59 00:03:58,460 --> 00:04:00,980 Lastly, we have ratings CFC. 60 00:04:01,250 --> 00:04:07,310 So, so far we looked at data about movies, but here we have data about users. 61 00:04:07,610 --> 00:04:13,400 So we are basically simulating a real world scenario here in real life. 62 00:04:13,400 --> 00:04:18,470 We would have this websites with movies, let's say Netflix. 63 00:04:19,660 --> 00:04:22,280 Working in Netflix as a programmers. 64 00:04:22,300 --> 00:04:25,080 We have databases, we have access to a database. 65 00:04:25,090 --> 00:04:26,620 We know about users. 66 00:04:26,620 --> 00:04:36,100 So, for example, we know that the user with ID one gave a rating of 2.5 to movie with ID 31, and 67 00:04:36,100 --> 00:04:38,590 that's the date of the rating. 68 00:04:38,710 --> 00:04:40,810 So a timestamp basically. 69 00:04:41,050 --> 00:04:45,130 So those are the ratings of user with ID one. 70 00:04:45,130 --> 00:04:47,860 And then we have another user with ID two. 71 00:04:47,890 --> 00:04:53,860 That user gave a rating of four to movie with ID ten, right? 72 00:04:54,760 --> 00:05:03,640 And so we will use this database in conjunction with the other two databases here to do collaborative 73 00:05:03,640 --> 00:05:04,510 filtering. 74 00:05:04,510 --> 00:05:08,360 So to build a recommendation system based on collaborative filtering. 75 00:05:08,380 --> 00:05:11,200 And yeah, these are the three datasets we have. 76 00:05:11,690 --> 00:05:18,740 So getting familiar with the data as we just did, here is the first step with any data science and 77 00:05:18,740 --> 00:05:20,590 machine learning projects. 78 00:05:20,600 --> 00:05:27,800 With that step completed, we're ready to go into Python and load this data there and then we can do 79 00:05:27,800 --> 00:05:29,300 some more exploration. 80 00:05:29,300 --> 00:05:31,400 So so let's do that in the next video. 81 00:05:31,400 --> 00:05:31,940 I'll see you there.