In a previous blog post
I demonstrated how the vector datatype in Elasticsearch can be used to search words by their semantic meaning.
In this post I will show how a reverse image search for paintings can be implemented using the same methods.
Given a photo of a painting, we will use Elasticsearch to find other paintings which look similar.
Have you ever wondered why IntelliJ can't resolve properties in your application.properties or application.yml?
Short answer: you'll need an annotation processor in your classpath.
Full-text search is one of Elasticsearch's strengths.
However, what if we wanted to find similar documents based on something more abstract - like the meaning of a word or the style of writing?
This is where Elasticsearch's dense vector field datatype and script-score queries for vector fields come into play.
Data scientists often find themselves spending a lot of time with data acquisition and preparation,
yet most tutorials start with ready to use datasets.
This time we will start with nothing but a simple problem and gather the data with scrapy to provide insight into the
process from data gathering to model creation.
Text classification is a common task where machine learning is applied.
Be it questions on a Q&A platform, a support request, an insurance claim or a business inquiry - all of these are usually written in free
form text and use vocabulary which might be specific to a certain field.
This article demonstrates how such classification problems can be tackled with the open source neural network library Keras.