10 Ways to Use Embeddings for Tabular ML Tasks

What’s Happening

Real talk: Embeddings — vector-based numerical representations of typically unstructured data like text — have been primarily popularized in the field of natural language processing (NLP).

10 Ways to Use Embeddings for Tabular ML Tasks By Iván Palomares Carrascosa on in Practical ML 0 Post Ways to Use Embeddings for Tabular ML Tasks Image by Editor Introduction Embeddings — vector-based numerical representations of typically unstructured data like text — have been primarily popularized in the field of natural language processing (NLP). But they are also a powerful tool to represent or supplement tabular data in other ML workflows. (plot twist fr)

Examples not only apply to text data, but also to categories with a high level of diversity of latent semantic properties.

The Details

This article uncovers 10 insightful uses of embeddings to use data at its fullest in a variety of ML tasks, models, or projects as a whole. Initial Setup : Some of the 10 strategies described below will be accompanied code excerpts.

An example toy dataset used in the examples is provided first, along with the most basic and commonplace imports needed in most of them. Import pandas as pd import numpy as np # Example customer reviews’ toy dataset df = pd.

Why This Matters

DataFrame(( “user_id”: [101, 102, 103, 101, 104], “product”: [“Phone”, “Laptop”, “Tablet”, “Laptop”, “Phone”], “category”: [“Electronics”, “Electronics”, “Electronics”, “Electronics”, “Electronics”], “review”: [“solid battery”, “fast performance”, “light weight”, “solid build quality”, “insane camera”], “rating”: [5, 4, 4, 5, 5] )) 1 2 3 4 5 6 7 8 9 10 11 import pandas as pd import numpy as np # Example customer reviews’ toy dataset df = pd . DataFrame ( ( “user_id” : [ 101 , 102 , 103 , 101 , 104 ] , “product” : [ “Phone” , “Laptop” , “Tablet” , “Laptop” , “Phone” ] , “category” : [ “Electronics” , “Electronics” , “Electronics” , “Electronics” , “Electronics” ] , “review” : [ “solid battery” , “fast performance” , “light weight” , “solid build quality” , “insane camera” ] , “rating” : [ 5 , 4 , 4 , 5 , 5 ] ) ) 1.

This adds to the ongoing AI race that’s captivating the tech world.