Semantic vector search is one of the newest developments, which brings the power of the latest improvements in artificial intelligence to known use cases in the field of search. Generally, vector search can help you find more relevant search results. How does this work in detail? For that matter, freely available language models are trained on large text corpora. These language models are then used to produce vectors of search queries and vectors of documents that encode meaning of words. As soon as the meaning is encoded, it can be matched to other words with a similar meaning. To sum up the process of optimized data search it is possible to find documents with similar words or meaning by using encoded and matching words in search queries. Thus, when successful, you can get more relevant results and avoid very few or zero results during your data search. The presentation will also show use cases from e-commerce to enterprise search as well as present open-source technologies that can be used for setting up a vector search solution.
Semantic vector search as newest development of AI in search
Broad use cases from e-commerce to enterprise search
Current open-source technologies for setting up a vector search solution
Project Manager, Solution Engineer
Bastian Mathes studied computer science and works as a project manager, solution engineer and architect. His focus is on building scalable search systems for their customers in different domains with open-source and commercial technology. Philipp Sieg studied linguistics and works as a project manager, solution engineer and computational linguistics expert. He focuses on multilingual search systems for customers in different domains, using open source and commercial solutions.
* Required fields.
Feel free to request recordings of past webinars. Please take a look at our recorded webinars on the overview page.
