Data Scientist / Machine Learning Exercise
- or -
Post a project like this$$$
- Posted:
- Proposals: 14
- Remote
- #2117554
- Expired
Digital Transformation | Advanced Analytics | Business Intelligence | AI | ML | Blockchain | AI-powered Chatbot
Thang
Trusted digital transformation partner of global enterprises. 14+ years of joyful journey with 450+ employees.
Bay Minette
234522322843258775135222819624052342306137182735586345782066042814226502344233
Description
Experience Level: Expert
We are looking for a data scientist / machine learning expert to work
We have 2 sets of data that have say "Product Names" and "vendors" on them of which we want to match against another list with slight variations on there i.e remove Inc, limited, underscores etc. So this you could argue is a fuzzy logic matching exercise but we would like the algorithm trained to work out what combinations of rules need to happen to allow one data set be matched against the other.
Where we believe this is a great candidate for machine learning is that we have the results to train the data against, i.e we have the results we want to match on which should be deemed successful so that the training algorithm is able to run with a supervised aspect. You also will be able to see visually what constitutes as a match and be able to tune your algorithm.
The data i pretty well laid out already so should be easy enough to organise. Its about 20,000 product names on 1 data set and about 50 on the other. There is about 600 matches between the 2 (the reason there is so many matches is that the product numbers have version numbers which match pretty well so 1 piece of software might have multiple version matches).
-I will issue data sets once we have chosen our person and have an NDA in place. its not super secret but we are doing this on behalf of a client.
-Would love to know in your response if you have done any fuzzy matching logic before
-We would require you to handover the knowledge once this POC is complete i.e overview of tensor flow / aws / azure tools (whatever you use).
-This is likely the first piece in a long line of work like this
-We are not really price sensitive here looking for someone that is smart, capable, good communicator.
-Please dont reply if you are not adept at Data Science / ML / AI type stuff no chancers please.
Thanks :)
We have 2 sets of data that have say "Product Names" and "vendors" on them of which we want to match against another list with slight variations on there i.e remove Inc, limited, underscores etc. So this you could argue is a fuzzy logic matching exercise but we would like the algorithm trained to work out what combinations of rules need to happen to allow one data set be matched against the other.
Where we believe this is a great candidate for machine learning is that we have the results to train the data against, i.e we have the results we want to match on which should be deemed successful so that the training algorithm is able to run with a supervised aspect. You also will be able to see visually what constitutes as a match and be able to tune your algorithm.
The data i pretty well laid out already so should be easy enough to organise. Its about 20,000 product names on 1 data set and about 50 on the other. There is about 600 matches between the 2 (the reason there is so many matches is that the product numbers have version numbers which match pretty well so 1 piece of software might have multiple version matches).
-I will issue data sets once we have chosen our person and have an NDA in place. its not super secret but we are doing this on behalf of a client.
-Would love to know in your response if you have done any fuzzy matching logic before
-We would require you to handover the knowledge once this POC is complete i.e overview of tensor flow / aws / azure tools (whatever you use).
-This is likely the first piece in a long line of work like this
-We are not really price sensitive here looking for someone that is smart, capable, good communicator.
-Please dont reply if you are not adept at Data Science / ML / AI type stuff no chancers please.
Thanks :)
Terry L.
97% (23)Projects Completed
26
Freelancers worked with
27
Projects awarded
34%
Last project
8 Feb 2023
United Kingdom
New Proposal
Login to your account and send a proposal now to get this project.
Log inClarification Board Ask a Question
-
There are no clarification messages.
We collect cookies to enable the proper functioning and security of our website, and to enhance your experience. By clicking on 'Accept All Cookies', you consent to the use of these cookies. You can change your 'Cookies Settings' at any time. For more information, please read ourCookie Policy
Cookie Settings
Accept All Cookies