Artificial Intelligence in Search Training
In this training you will learn how to design and implement intelligent natural language search leveraging Open Source software.
You will learn about language modelling, vector based search and how to integrate it with Apache Lucene/Solr and Elasticsearch.
Would you like to schedule a training for your Team?
Our trainings are meant for different targets ( business, developer and everything in between), and are individual/small groups/classroom-based.
We offer different levels of flexibility to make sure the training course will be a perfect fit for your requirements
ON SITE OR REMOTE
LIVE OR RECORDED
FROM BEGINNER TO EXPERT
Skills You Will Gain
• How to integrate natural language processing techniques with your search engine
• How to use pre-trained language models and fine tune them for your specific use case
• The pros/cons of vector based search
• How to do that with Apache Lucene/Solr and Elasticsearch
Basic understanding of Search Engines and Machine Learning
Software Engineers, Data Scientists, Machine Learning passionates.
APACHE LUCENE/SOLR COMMITTER
Alessandro has been involved in designing and developing search-relevant solutions from 2010.
Over the years he has worked on various projects, with various open source technologies aiming to build search solutions able to satisfy the user information needs, often integrating such solutions with machine learning and artificial intelligence technologies.
Natural Language Processing
NLP tasks useful for Search
Open Source Libraries Overview
Text and Speech Processing(Optical Character Recognition, Speech Recognition, Text to Speech, Tokenization)
Morphological Analysis(Stemming/Lemmatization, Part Of Speech Tagging)
Syntactic Analysis(Sentence Breaking)
Lexical Semantics(Named Entity Recognition and Linking, Sentiment Analysis, Word Sense Disambiguation)
Implicit Semantic(Topic Segmentation and Recognition)
Higher Level Application(Automatic Summarization, Grammatical Error Correction, Machine Translation, Natural Language Understanding)
- Hands on Exercises
Deep Learning for Search
- A bit of history (challenges with exact matching, Deep Learning, semantic search)
- Word Embeddings
- Hands on Exercises: let’s start generating and using word embeddings
- BERT and transformers
- Hands on Exercises: let’s use BERT
- Model fine-tuning (pairwise learning, Margin-MSE aka DistilBERT)
- Hands on Exercises: Fine-tuning with pairwise learning
- Similarity search of dense vectors
- Hands on Exercises: Perform ANN search
- Augmenting the inverted index (document expansion and emphasize important terms)
- Hands on Exercises: performing document expansion
Vector Based Search
- How Open Source Search Engines cover the functionality (Apache Lucene/Solr, Elasticsearch and Vespa)
- Hands on Exercises: let’s run Vector-based search in Open Source search engines
- Query Intent Classification
- Hands on Exercises: recognise the user intent with Open Source
- Zero result queries
- Hands on Exercises: improvement for zero result queries with Open Source backed techniques
- Measure your progress
- Explainable artificial intelligence
The positive feedback I had about your professionalism and skills in the areas of information retrieval and AI makes me think that you can definitely be an important partner for the growth and development of our products."
Where to start
If you wish to be introduced to Artificial Intelligence applied to Search, we have written a series of blog posts about it, that may be a good start to this subject.