Example applications

Here are some example applications that use similarity search and Pinecone.

Our Learn section explains the basics of vector databases and similarity search as a service.

How to create a simple semantic text search using Pinecone’s similarity search service.

Semantic Textual Search

View Source

Open In Colab

Pinecone allows you to pair semantic search with a basic keyword filter. If you know that the document you're looking for contains a specific word or set of words, you simply tell Pinecone to restrict the search to only include documents with those keywords.

Basic Hybrid Search

View Source

Open In Colab

Question-answering

How to build a question-answering application with similarity search and Pinecone.

We will walk you through how to index a set of questions and retrieve the most similar stored questions for a new (unseen) question. That way, you can link a new question to answers you might already have.

Question Answering

View Source

Open In Colab

How to perform time-series "pattern" matching using a similarity search service.

In this example, we will explain what to do if you wanted to retrieve all the historical time series items that matched a particular pattern.

Time Series Similarity Search

View Source

Open In Colab

Video recommendations

How to create a movie recommendation system on the Movielens dataset.

Video Recommendation

View Source

Open In Colab

How to use Pinecone’s similarity search as a service to build an audio search application.

This enables you to:

  • Find songs and metadata within a catalog, based on a sample
  • Find similar sounds in an audio library
  • Detect who's speaking in an audio file
  • Take some new (unseen) audio recordings and search through the index to find the most similar matches, along with their YouTube links.

Audio Similarity Search

View Source

Open In Colab

Document deduplication

How to create a simple application for identifying duplicate documents.

Document Deduplication

View Source

Open In Colab

How to label new texts automatically when there is an enormous number of potential labels.

This scenario is known as extreme classification. This is a supervised learning variant that deals with multi-class and multi-label problems involving many choices.

Extreme Classification

View Source

Open In Colab

How to create an image similarity search backend service.

You will learn how to:

  • Use the pre-trained embedding model called squeezenet from torchvision to transform image data into vector embeddings
  • Build an index with Pinecone to store these vector embeddings
  • Send a new image as query, and retrieve similar images in the index

Image Similarity Search

View Source

Open In Colab

IT threat detection

How to use Pinecone similarity search to build an application for detecting rare events.

Such application is common in cyber-security and fraud detection domains, wherein only a tiny fraction of the events are malicious.

IT Threat Detection

View Source

Open In Colab

Personalized article recommender

How to use Pinecone's similarity search to create a simple personalized article or content recommender.

Personalized Article Recommender

View Source

Open In Colab