Click here to Skip to main content
15,795,318 members
Please Sign up or sign in to vote.
0.00/5 (No votes)
Need guidance on an ML/NLP project for a newbie!
Hey guys, I'm sorry if this is not the kind of thing that's asked in this website.

Me and 3 of my friends from uni were given a project, a challenge project to be done in 2 months(around 57 days left now) to do a project where we analyze customer reviews on restaurants(content based filtering where we analyze their reviews and identify food and break food down into ingredients?), and we predict recipes to them based on given reviews after processing the review text.

For this, we decided on using the Yelp challenge dataset @

We also have a different database for recipes.

My question is, how would one actually go about developing this system? For newbies like us, it doesnt really look easy, but we are willing to put in a lot of time and effort for this.

We all have a good knowledge in coding Python and Java.

How would you distribute the work among 4 people? Like, one person does NLP, one person does ML, one person does neural networks or something?
We have literally no background on these technologies and that's why Im here, asking the experts. Please could someone give a detailed overview on how we could go about doing this in such a short amount of time?

Thanks in advance guys, and Im sorry again if this is the wrong type of thing to ask here. I'm not asking anyone to code for me, I just need some guidance. Thanks again!

What I have tried:

Well, we've done the text extraction and natural language processing to filter and preprocess the data, but that's just about it.
Updated 25-Feb-19 0:29am

You could take a look at these open source GitHub projects to see how they do it: Top 20 Python AI and Machine Learning Open Source Projects[^]

And here is a Python project template developed by the University of Washington: GitHub - uwescience/shablona: A template for small scientific python projects[^]

You can also use a Project Management tool, see overview here:[^]
Share this answer
If you do decide to distribute one topic per person then do make sure the work load is evenly balanced and that everyone is at the same level of understanding and ability.

So, however you decide to do it in the end,
- Get together often,
- brainstorm your solutions together often,
- be supportive to each other - all the time

Once you have a pen picture of how you want to proceed, you could also run your ideas for managing the project past your tutor - just make sure you are going to them with a potential solution and not just asking for help (having read this question I'm sure you would do that anyway)

Another technique that I've used in the past (it is a bit "old school") is to buddy up - so pair up on one specific topic (You can mix and match the pairs as the project progresses) - bouncing ideas around with someone else can often be better than struggling alone.

I let the side down on a similar project nearly 40 years ago, Yes, I still carry the guilt. Good Luck!
Share this answer

This content, along with any associated source code and files, is licensed under The Code Project Open License (CPOL)

CodeProject, 20 Bay Street, 11th Floor Toronto, Ontario, Canada M5J 2N8 +1 (416) 849-8900