Hi, I am Sandeep. I made a transition from working as a Senior Software Engineer with Bank of America for 6 years to the field of data science and analytics. I belive that you need to know the data/domain before you deliver things of value to your users/clients.
The two quotes below help me understand the data better:
- There are three kinds of lies - lies, damned lies and statistics.(Mark Twain)
- Understanding of life begins with the understanding of patterns.(Fritjof Capra)
At times, we are so busy thinking about models and metrics we fail to understand the problem and the data. Here is my view on the approach one should take before jumping into modelling.
CarDekho.com is India's leading car search venture that helps users buy cars that are right for them. One of the useful information for the users would be to know the current price of used/new cars.
My Goal from this project is to understand the challanges of end to end machine learning workflow. Takeaways from the project
As we are dealing with large volume, velocity, and variety of data, the veracity, and the value we derive from it has been of prime importance. False information is easy to spread and can have a significant impact on genuine people or customers. Like other sectors, the job market has also been impacted by it.
The state of art NLP technique developed by Google known as Bidirectional Encoder Representations from Transformers (BERT) is being used for this problem. BERT helps better understand the nuances and context of words in searches and better match those queries with more relevant results.
Use Amazon Product Co-purchase data to make Book Recommendations using Network Analysis.
Please click on the header for the methodology description and the below links for code & data.