I am looking for someone who can develop a python script (preferably in Jupyter notebook) that can calculate the Euclidian distance of records in a dataset and select the K nearest neighbor record based on the smallest Euclidian distance.
1. Dataset is large with over 550,000 records
2. Script will import the dataset from a .xls/csv file.
3. A weighted Euclidian distance will be calculated based on a defined weight placed on 5 variables in the data set. 4 of the variables are nominal (i.e. Site IDs, or part number) and therefore the distance of one of these nominal variables between records would be a 1 = no match, 0=match.
4. For each record, the K nearest neighbor record (i.e. smallest weighted Euclidian distance) will be identified. The actual Euclidian distance will also be referenced.
5. The dataset with the additional fields for K-nearest record's id and its Euclidian distance can be exported to excel.
6. # Notes describing each step should be referenced in the python script.
I have attached a file with a sample of the data set (sheet 1). Yellow fields are the variables that are used for the distance calculation between records. On Tab 2 the weights for each variable to apply in the Euclidian calculation are referenced.
If you are interested, please respond with the following:
a) Your fixed price quote for development of the python script.
b) Date when you could have the script developed. (needed asap)
c) Any questions you have regarding the request.
14 freelancere byder i gennemsnit ₹13196 på dette job
HI..i am proficient in python programming and familiar with k-nearest neighbor algorithm and can help you implement the code in Python using Jupyter notebook using python 3.
K nearest points with calculation of the Euclidean distance was part of my Master's thesis. I think I can easily perform the task with an efficient solution.
Hi there ! We are group of graduate students of Data Science. We can perform your project and it will be done in a Jupyter notebook. The project will be delivered to you within 4 days i.e before Friday. Regards.