Build A Movie Recommendation Engine Using Python
ฝัง
- เผยแพร่เมื่อ 23 ก.ค. 2024
- Build A Recommendation Engine Using Python
⭐Please Subscribe !⭐
⭐Get the code and data sets by becoming a supporter on Patreon:
/ computerscience
⭐Websites:
► everythingcomputerscience.com/
⭐C-Programming Tutorial:
► www.udemy.com/course/c-progra...
⭐Supporter the channel on Patreon:
► / computerscience
⭐Helpful Programming Books
► Python (Hands-Machine-Learning-Scikit-Learn-TensorFlow):
amzn.to/2AD1axD
► Learning Python:
amzn.to/3dQGrEB
►Head First Python:
amzn.to/3fUxDiO
► C-Programming :
amzn.to/2X0N6Wa
► Head First Java:
amzn.to/2LxMlhT
#Python #RecommendationProgram #RecommendationEngine - วิทยาศาสตร์และเทคโนโลยี
Great tutorial.
In order to help others, the dataset can be found by searching "IMDB data from 2006 to 2016" and you just need to add the Movie_id column. That can be easily solve by running something like: df['Movie_id'] = range(0,1000) right after importing the data, or simply update the csv.
Thanks boss
hi there, can u help me with one thing pls
Can u tell how can i check if the string is similar to movie name or not because this program that he have coded only recommends if the movie name is as same as the csv dataset but if someone only types in spiderman instead of full correct title then it will show an index error, which i have kept as an except for now but how can i check if the string is similar to title or something in csv dataset
Clutch
awesome video, exactly what i was looking for, had a few errors but it was worth going through the whole video. Thank you👍
This is really great. After going through lot many links and blogs got something really useful. My understanding of cosine similarity is clear now.
You can't learn about cosine similarity with this video -it is used just as a black box. This gives more insight to the basic math behind cosine similarity: www.kdnuggets.com/2019/04/building-recommender-system.html
Cool, thanks Computer Science.
Nice that it's simple, very doable.
as someone who is just getting started with coding i really enjoyed watching this. thank you for your work . this is great ^_^
Thank you! Im glad this video could help!
I don't know how I should to appreciate you but the most important thing is that your work is really appreciatable
Good information, with clear sound and simple steps ,thanks for sharing 👍
Good tutorial, I learned a decent amount! I'm wondering what the next layer down is from here, however. Is there a way to weigh certain factors more heavily? Like title is a good feature, it's an easy way to detect other movies in a franchise, but the similarity scores seem very, very low for anything that doesn't share a similar title. If I want to make it so that description or genre or actors can have adjusted values, is that a feature?
That was si useful. Thank you.
Where we can get the dataset that you mentioned in the video?could you please tell us where it is available
Hey, I loved your video it explained a lot to me. But I have a question when you implement this in the real world. I can see that feed the data from an cvs file, like all the tutorials online. In real life you probably have a database with new movies all the time, what happens when you input a new data, you have to run this script to generate the similarity matrix? Does it become cpu expensive as your database grows?
You should use the str accessor of the dataframe to join columns (when you created that combined important features column). Using a for loop on a dataframe is neither convenient nor efficient.
very good work. but please I have confusion for performance evaluation metric of recommendation system like rank10, precision, recall etc.thanks
Great video! How can i make the model to take in multiple input of titles instead of 1? and then the recommendation will be based on all the multiple inputs
Can u tell how can i check if the string is similar to movie name or not because this program that you have coded only recommends if the movie name is as same as the csv dataset but if someone only types in spiderman instead of full correct title then it will show an index error, which i have kept as an except for now but how can i check if the string is similar to title or something in csv dataset
great content, can u make a tutorial to find out the corellation between the features in the csv data ?
Bro, u r the best. Thank u a lot!!!
Great TUT easy to understand
like from India
Wow this is amazing
what to do incase user lkes more than one movie like input is spiderman and iron man then how should we go about using content based filtering
What if we want to campare the similarity of the movie which is not in the database but know its attributes
Wow , pretty impressive.
hey what type of filtering is this?
podrian agregar el enlace de descarga del archivo cvs por favor, es que no lo encuentro en ninguna pagina. Gracias
what does it mean?? I'm having problems on [9] #Create a column to hold the combined string.
it shows this: ValueError: Length of values (1) does not match length of index (1000)
I'm getting the same error please help
Did anyone figure out how to solve this error?
Can someone please share the 'movie_data.csv' file or its source?
Help me
The error say : TypeError: 'int' object is not subdcriptable in finally video or code in show the recomendation
Adamsın adam gibi adam adam
Can I ask you what algorithm you use?
so is this a machine learning model or just algorithm ?, cuz i dont see any training process
What algorithm do you use?
Sir if we add new movie in the website, we need to add the new movie to the dataset and create model again
I'm starting and this video helped me a lot, however how can I display more than one info in the final result (for example the name of the movie and the date of publication). Thanks in advance
Hi sir what is the difficulty of this tutorial? i mean this is the basic of ml or advanced?
Great work! how can we implement it on an android app? can we convert it into TF Lite?
I'm getting an error at the part where we do cm = CountVectorizer().fit_transform(df['important_features']) Please help me
if it says that object does not have a lower method or something like that, you should make the important_features column as a string, so you can do df['important_features'] = df['important_features'].astype['str']
@@alicerezenderibeiro1896 thanks
IndexError Traceback (most recent call last)
in
1 #Create a list of enumeration for the similarity score[ (car price,similar cars)....]
----> 2 scores = list(cs[id])
IndexError: only integers, slices (`:`), ellipsis (`...`), numpy.newaxis (`None`) and integer or boolean arrays are valid indices
i tried same for car recommender but getting this error at list enumerate any help ?
getting the same error :(
can we export this to be a model ?
hey!! i getting error in DF['important_features']=get_important_features(DF) and error is KeyError: 206 can u plz help me solve this..!!!
Try this out:
df['important_features'] = df['column_1']+' '+ df['column_2']+ ' ' ....+ df['column_n']
I think you should get the same result.
There is no need to create a function and iterate using for loop.
Im getting an error in this part:
df['features'] = get_important_features(df)
df.head(30)
try this in your code...had the same error and was able to solve with this...
#create a funtion to combine the values of the important columns into a single string
def get_important_features(data):
important_features = []
for i in range(0, data.shape[0]):
lis = [data['type'][i], str(data['title'][i]), str(data['rating'][i]),data['genres'][i]]
x = ' '.join(lis)
important_features.append(x)
return important_features
@@hamisiandale2942 You are a life saver. Thank you
Why don't you come up with something new..you are really one of talented people out there..
I m getting error in cm = CountVectorizer line ,
'Series' object has no attribute 'lower'
Plz resolve this error
so do I
@Shrimad Mishra I m getting error in cm = CountVectorizer line ,
'Series' object has no attribute 'lower'
Plz resolve this error
Sir is this content based or collaborative filtering method?
Did you get your answer?
Content based
I think adding input function for user to input the movie title can be much better
Can you give the movie database
Where did the data come from?
Google the attributes in the order as given The first result gives you the result the data set in kaggle
where to find the data sset
IMBD-Movie-Data kaggle
its a good program can you please tell us where did you get movie data from ?
You can download data sets on keggle for free :)
@@HankeyMassacre yoooooooo i was searching soooo much to know how to get the data your reply just saved me thanks
please upload your csv file in description
Can you please upload a video on music recommendation using CNN
That would be interesting, I will definitely consider it.
from vietnam with love
Thanks, I truly appreciate it !
I'd like to clarify, this is not really a "recommendation" system comparing to Netflix or Amazon Video, it is more like a "similar movie search" feature, which can be part of the search results when you search for movies. It is still a useful thing, but if someone is looking to build a recommendation system, especially when the "important feature" is not yet clear, this video would likely give them a wrong understanding. A recommendation system should at least have a user preference table and usually uses matrix factorization to figure out the important features. I'd suggest change the title to be more clear for the purpose of this video...
Thanks for your feed back, I truly appreciate it. I will not change the title of this video however and I will explain why.
The code in the video is indeed a recommendation engine/system by definition. According to wikipedia “Recommender systems or recommendation systems (sometimes replacing “system” with a synonym such as platform or engine) are a subclass of information filtering system that seek to predict the ‘rating’ or ‘preference’ that a user would give to an item.”
The recommendation engine in this video is a content based recommendation engine. A content based recommender works with data that the user provides, either explicitly (rating) or implicitly (clicking on a link).- Analytics Vidhya
In the video we get the movie that the user likes and then recommend similar movies.
@@ComputerSciencecompsci112358 I am trying to build a system similar to TH-cam's recommendation system will this be good?
How to change into web app
Upload code n data set
I need the code file please anyone help me ?
Hi PanicMod,
You can get all of the code and the dataset at patreon.com/computerscience
@@ComputerSciencecompsci112358 i am a student dont have a credit card can you give this to me free ?
@@PanicMod you can try google 'IMDB data from 2006 to 2016' as recommended from top comment if you still looking.
I'm getting error on df[columns].head(3) can anyone help me out please
what's the error?
@@alrifat9419 I just told you
I am getting the error below please help:
ValueError Traceback (most recent call last)
in ()
1 #Create a column to hold the combined strings
----> 2 df['important_features'] = get_important_features(df)
3
4 #Show the data
5 df.head(3)
3 frames
/usr/local/lib/python3.7/dist-packages/pandas/core/internals/construction.py in sanitize_index(data, index)
746 if len(data) != len(index):
747 raise ValueError(
--> 748 "Length of values "
749 f"({len(data)}) "
750 "does not match length of index "
ValueError: Length of values (1) does not match length of index (1000)
Indent the above return statement with the for loop statement
Thanks for video , but its better to explain its content-base ? user- base ? what kind of recommender system ?!
Did you get your answer?
@@vinaybora617 not important anymore i created my own recommendation system , and defended my master thesis with excellent mark !
@@Bawaromerali great 👍
@@vinaybora617 thanks
#get the title of the job that the user likes
title = 'IT Security Data Scientist'
#find the job description
job_description = df[df.Job Title == title]['Job Description'].values[0]
this is the 16th code set and I get a syntax error calling Invalid Syntax and the error points out the Job Title and I still couldn't figure it out
Can I convert this to .exe?
what do you mean by converting it to .exe? do you mean if you want to build an app ?
@@armitosmt5753 yes an app
So, to be clear, this is not a machine learning model, right?
Please can you give me the full code of this recommender engine
How to drop users with less than 10 reviews
Teh
** Urgent Please Error
---------------------------------------------------------------------------
IndexError Traceback (most recent call last)
in ()
3
4 #Find the movies id
----> 5 movie_id = df[df.movie_title == title]['Movie_id'].values[0]
IndexError: index 0 is out of bounds for axis 0 with size 0
Your content is really good, the workflow is smooth, however, cannot implement them on our own because you provide no links for the datasets. Such a waste.