Search with Machine Learning
This course covers the fundamentals of integrating machine learning and natural language processing techniques into search engines. We'll learn how to use machine learning for ranking, content understanding, and query understanding, along with how to use embeddings, dense vectors and deep learning to improve retrieval and ranking. You will build applications using OpenSearch (an open fork of Elasticsearch) and several ML libraries and plugins. Note: This course assumes that you have already learned search fundamentals; if not, then we encourage you to take our companion “Search Fundamentals” course so that you will be prepared for this one.
Course taught by expert instructors
Former CTO at Wikimedia
Grant is a CTO, independent consultant and advisor. He is the former CTO of the Wikimedia Foundation and the co-founder and ex-CTO of Lucidworks, co-author of Taming Text, co-founder of Apache Mahout and a long-standing committer on the Apache Lucene and Solr open source projects. Grant’s experience includes managing a large team of engineers, researchers and data scientists at a top ten website as well as engineering a variety of search, question answering, and natural language processing applications for a variety of domains and languages. He earned his B.S. from Amherst College in Math and Computer Science and his M.S. in Computer Science from Syracuse University.
Machine Learning Consultant
Daniel is an independent consultant specializing in search, machine learning / AI, and data science. He was a founding employee of Endeca, a search pioneer that Oracle acquired in 2011. He then led engineering and data science teams at Google and LinkedIn. He’s worked with a wide range of consulting clients, including Apple, eBay, Pinterest, Salesforce, Yelp, and Zoom. He wrote a book on Faceted Search, published by Morgan & Claypool, and he blogs on Medium about search-related topics — particularly query understanding. Daniel has degrees in Computer Science and Math from MIT and a PhD in computer science from CMU.
Learn and apply skills with real-world projects.
Software engineers and data scientists looking to learn more about how search engines work, and specifically where they can best apply machine learning to improve search quality.
- Machine learning experience is NOT required.
Ability to write Python and work with documented libraries.
Comfort working with web applications, Docker basics (e.g. start, stop) and the command line.
Search Fundamentals (https://corise.com/course/search-fundamentals) class certificate or academic or industry experience working with search engines such as Elasticsearch/OpenSearch/Solr/Vespa. We will not be teaching search fundamentals in this course.
Try these prep courses first
You will create, deploy, and experiment with a machine-learning ranking model using ecommerce application data and query logs.
- How to model and measure search relevance
- Relevance, ranking, diversity – how they all fit together
- How to use machine learning for ranking results
You will enrich the ecommerce application from weeks 1 and 2 with content attributes derived from rules-based and machine-learned classification, as well as using classifiers to automatically enrich new unlabeled content.
- Techniques for content classification and annotation, such as entity recognition
- Supervised and unsupervised machine learning methods for content understanding
- How to build, evaluate, and improve a neural content classifier
You will enrich the ecommerce application from previous weeks with machine-learned query understanding, based on query-content-label triples, to improve retrieval and relevance.
- Fundamentals and key components of query understanding
- How to combine query understanding with retrieval and ranking
- How to build, evaluate, and improve a neural query classifier
The inverted index has served as the foundation of most search engines for decades, and still does so today. But the emergence of word embeddings has been driving a semantic search revolution driven by vector representations and nearest-neighbor methods. In this project, you will use embeddings to represent products and queries as vectors and explore vector search approaches for retrieval and ranking.
- Using dense vector representations for semantic search
- Indexing and querying approaches for dense vectors
- Using vectors for query and content similarity
Work on projects that bring your learning to life.
Made to be directly applicable in your work.
Live access to experts
Sessions and Q&As with our expert instructors, along with real-world projects.
Network & community
Core reviews a study groups. Share experiences and learn alongside a global network of professionals.
Support & accountability
We have a system in place to make sure you complete the course, and to help nudge you along the way.
Course success stories
Learn together and share experiences with other industry professionals
I feel that taking the course helped me better understand the Search domain in my company, which is very useful for my work. Thanks for a great experience!
Best online course I've taken!
The best way to learn how to apply ML to search in a collaborative and friendly way !
If you want to learn practical knowledge of using Machine Learning on Opensearch, this course is definitely for you.
It was difficult, but a level of difficulty that will make it proud of yourself! I learnt so much! Exactly what I dreamt of.
I had a blast! 4 weeks of intense immersion, with great instructors, and a very open and sharing community I now have new tools in my toolbox. And new additions to my professional network to ask when the need arises.
I have been working in Search for a long time and 90% of this time I have used Solr / Lucene. I wanted to learn about other search engines so I enrolled in this course. I have delivered Solr trainings myself in the past and I will continue in the future too, and my participation in this course will certainly change my training style! I loved the insightful questions and discussions and some questions helped me open up my perspectives about certain concepts. Big thank you to both the trainers Grant and Daniel and organizers -- it was an awesome class and best wishes to the future classes!
A must have course for search engineers and anyone looking for a solid introduction to search with ML. Highly recommended!
The course is really great if you're someone who likes to work hands-on first and learn a long the way as you're doing the assignment. The time schedule is very flexible, lectures are recorded, multiple coding sessions for different time zones. It'd say this is one of the best teaching approaches I've tried so far!
Search with Machine Learning is an incredible course taught by two industry experts and a community of fellow search engineers. It was an intense four weeks but I walked away with a deep understanding of how to build a great search engine. I am already applying these skills in my workplace and will surely leverage this knowledge in the future.
This course is a great resource. It's like a live reference book for anything from Indexing to Query Understanding. The authors are very approachable and they share a lot of their expertise on the techniques and their business applications. And it's good to see so many people working in the field and being so generous with their time to help and share their experience. Excellent overall!
The co:rise Search with Machine Learning class was four solid weeks of the right mix of lectures, eminent guests' talks, weekly project/homework, and on-going near real-time interactions with the industry veterans instructors as well as the co-students. Just that interaction was priceless! As a search engineer, consultant, and practitioner, the co:rise class significantly strengthened my knowledge of, and confidence in, the know-how of Learning-to-Rank, and the use of Machine Learning techniques for content and query classifications, all for improving Relevancy. I strongly recommend this class to anyone involved in designing, developing, and supporting information retrieval ("search") solutions. Thank you Grant, Daniel, Judy, and Amber at co:rise for a great class.
Thanks for the awesome course! Really helped me to improve my skillset in area of Information retreival.
I'm happy to have taken this course: it is well organized, with very supporting and caring teaching crew, Grant and Daniel. Both bring unique perspectives -- engineer's and data scientist's -- into the topic of Search, showing how multifaceted it has become over the years. The atmosphere on the course has been awesome, everyone is so supportive and sharing ideas and asking tough questions or even sharing that little recipe to overcome a programming / infrastructure issue. Saved me a ton of time! I've also enjoyed interacting with Judy during the course as she was timely checking in on how I am doing and what kind of feedback I've got, overall creating a positive and supportive vibe to keep going. I have already recommended the course to my clients. You will get a shared feeling of that you are not the only person in the world solving a tough Search challenge, and may be even meet new friends in the field.
The whole course was extremely relevant (no pun intended) for my work in search and I could feel the ideas coming up in my head as the weeks progressed.
This has been a great learning experience of applying some of the concepts in machine learning to search!
I can think of no better person to teach a class on 'search with machine learning' than Grant Ingersoll. Through his open source work on the Apache Solr search engine and as a founder of the Apache Mahout machine learning framework, Grant has done more to teach developers about the technology concepts and applications of ML and search than anyone in the business. If you type “machine learning “and “search” into Google, the top result is Lucidworks, a company that Grant was the driving force for. This is a no-brainer, take this class from Grant!
Grant is an established expert in the area of Search, Machine Learning, and AI! He has the mind of a researcher, an educator and has demonstrated the applicability of these deep topic areas into real-world products used by enterprises. What better way to learn than from someone who is a published author, a hands-on practitioner, and an industry expert?
Grant has a unique combination of breadth and depth in the search space, from designing a search solution for performance and stability at scale, while also having deep expertise in the internals of the engine, such as text analysis, relevancy tuning, query optimization, and index design. You won’t find a better teacher to provide a solid foundation in the theory as well as how to apply it when building real world search applications.
Daniel has unparalleled experience in search and machine learning. He brings practical experiences and has consulted for a wide range of companies from Apple to Zoom.
Daniel is a well-known expert on what useful search looks like in practice. Between his founding role as Endeca's chief scientist, his experience at Google and LinkedIn, and his consulting for numerous tech companies and retailers, he has seen and done it all. If you want to learn about relevance, user happiness, and optimizing search applications so they actually help people find things, you cannot do better than a course from Daniel.
For those with a good ML background and an interest in search: This is the best course I've found that is the correct balance between the two subjects.
This is brilliant course with a strong practical focus. It helped me a lot in building my understanding of successful search techniques used by industry. I was taking this course with my colleague and we build a search roadmap using learnings from the test.