Aditya Kumar

Ranking the results of ML models with a time decay factor for large-scale Anomaly detection

In this blog post, I will go through one of the problems I have recently faced that seems to be reoccurring in some form or another, which requires a ranking of the results produced by machine learning models in absence of implicit, explicit, or delayed feedback. I will explain the background, and problem and then discuss the approaches that can be used to solve the problem. And finally, how it can be easily implemented in any database and can be scheduled as a task in DB.

Monitoring Machine Learning Models in Production

After deploying many ML models in production, it became evident that there should be an easy and efficient way to monitor the ML models after deployment. This blog post is focused on monitoring the classification models in production.

Recently, I was working on the text classification problem which will classify the text into one of ~50 categories. Once the model is built and tested, it needs to be deployed as a flask API along with other models. Some text classification models are already deployed as an API that uses python flask to serve the incoming requests which use Gunicorn as a WSGI server and are deployed on Kubernetes clusters and trained models are stored in S3.

Wilson Lower bound Score and Bayesian Approximation for K star scale rating to Rate products

As a maintainer of an online community, which is having a lot of products where user gives a rating to products based on their experience, then it is definite that at some point you have to find an answer to questions like

• How you are going to show the product on the page based on filters i.e. like highest voted or lowest voted, etc.?

Maximal Marginal Relevance to Re-rank results in KeyPhrase Extraction

Maximal Marginal Relevance a.k.a. MMR has been introduced in this paper The Use of MMR, Diversity-Based Reranking for Reordering Documents and Producing Summaries. MMR tries to reduce the redundancy of results while at the same time maintaining query relevance of results for already ranked documents/phrases etc.

Deep Averaging Network in Universal Sentence Encoder

Word embeddings are now state of art for doing downstream NLP tasks such as text classification, sentiment analysis, sentence similarity etc. and provides very good results compared to tf-idf or count vectorizer. Using word embeddings we can find the similarity between words and can apply vector