Personalization! Everyone’s doing it, so is it even special anymore? Personalization has been a very hot topic these days which is being blended everywhere including search results as well. It means the search results are ordered to meet the individual needs of a consumer. Here, I will be taking you through the application of personalization for search in the eCommerce domain, also known as personalized search.

Image: Source

There are too many choices, you end up not knowing which one you want — Black Mirror

Search


All About Deep Learning

Training ANN and Data Preprocessing

We have talked a lot about the Feed Forward Neural Network in the last chapter. Now, we will talk about Tensorflow a bit and then, the implementation of FFNN in TensorFlow.

TF for DL


All About Deep Learning

Introduction

I guess people like to read about the things that they want the most and experience the least.Bernard Lowe (Westworld)

And guess what? embarking on this series. The series illustrates Deep Learning. ML / DL! Hmm. Actually, DL is not different from ML, but DL(Deep Learning) is a subfield of ML(Machine Learning) concerned with algorithms that are inspired by the structure and function of the brain known as Artificial Neural Network(ANN). As “neuron” is the basic working unit of the brain, it is also the basic unit of computation in a neural network.

Neural Network Architecture

We will straight forward discuss neuron…


E-commerce Search

As a continuation of the Search Query Understanding, here comes the next problem to solve which is known as Holistic Query Understanding. We will talk about this in detail as we have already covered the Reductionist Query Understanding part here already. Here, we are taking the case of e-commerce search. The problem is to find the intent of query lying in a particular taxonomy like L1/L2/L3 category also called category classification.


Microservices in recent days has been the very hottest topic in technology and the microservice architecture is being followed by technology giants like Netflix, Twitter, Amazon, Walmart, etc. as well as several startups. They are the perfect fit for the nowadays agile software development process where continuous innovation happens and products are continuously delivered. Let's find some more details about microservices:

The Basics of Microservices: What and Why?

Going by the definition:

In microservice architectures, applications are built and deployed as simple, highly decoupled, focussed services. …


In English Language, people generally type the queries which are separated by space, but sometimes and somehow this space is found to be omitted by unintentional mistake. The method of finding the word boundaries is known as Query Segmentation. For example, we can easily decipher

nutfreechocolates

as

nut free chocolates

But, the machine can’t unless we teach them. Here is a piece of evidence:

Showing results of an unsegmented search query

Since we have not lost our capabilities, our brain does this somehow intuitively and unconsciously. But, in the search domain, it affects the precision and recall of the results for a particular query which is not…


Deployment

Simply put, things always had to be in a production-ready state: if you wrote it, you darn well had to be there to get it running! — Mike Miller

We have containerized the application separately using docker, so as to keep it de-coupled. So, there are two images whose instances will communicate with each other — one for the Crocodile Model and another for the Service Orchestrator. Image for TF-Serving based Crocodile Model:

image: crocodile-model:latest
command: --model_config_file=/models/models.conf
environment:
MODEL_NAME: crocodile_model
ports:
- "8501:8501"
networks:
qus_net:
ipv4_address: 172.16.0.2

Image for Service Orchestrator:

image: service
ports:
- "8080:80"
volumes:
- ./log:/app/log


Machine Learning

“Is artificial intelligence less than our intelligence?” — Spike Jonze

Now, in this part, we will discuss the ML model and the deployment from the cloud infrastructure perspective.

ML Model, which we call [Crocodile Model] :) ,

Broadly, It can be divided into three parts:

1. Word Representation: First of all, we need to represent the words in the search query in the form of a feature vector, for which we have used pre-trained word embedding model developed…


Introduction

The ultimate search engine would basically understand everything in the world, and it would always give you the right thing. And we’re a long, long ways from that. — Larry Page

We call it Query Understanding (QU). Let’s first understand the use case through a problem. I assume myself to be a very health conscious person and want to buy a very healthy edible. I ended up searching for a bar of chocolate which should not contain nut as nutrition.

Normal Search Response

So, I searched for: “nut free chocolate”.

But, wait what I am seeing here is biscuits, coffee sachets, ginger nuts…

Sonu Sharma

Software Engineer @Walmart | Search | Linkedin: ercsonu

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store