haystack 2023

[manisnesan]

Youtube - Playlist

Aruna Lakshmanan gave an awesome Lightning talk with tons of in-depth advice around search signals. I thought it would have been even better as a full size talk or workshop. Here are a list of user signals she spoke about.

classify query term (brand/category/keyword, search vs landing, top product/category, keywords) facets (click order, facets missed) search vs features (don't load features up front) -- what are the top features that are being clicked? click metrics -- not clicked results? zero results and recommendations (should be based on user signals) time per session (longer) drop rate personalization, preference and trending

[manisnesan]

Lucene 9 vector implementation, by the folks at KMW Technology – Lucene and Solr 9.x support ANN search for vectors, but the index needs to be in a single segment and is loaded into memory in its entirety, making it not very useful for large vector indexes. Large indexes can be supported but at higher cost.

[manisnesan]

From Sujitpal - Slides

Question Answering using Question Generation

I didn't technically attend this talk since this was my presentation, but I was there in the room when it happened, so I figured that counts. In any case, this was my talk, its about the work I did last year with fellow data scientist Sharvari Jadhav to build a FAQ style query pipeline proof of concept using a T5 sequence to sequence model to generate questions from passages, storing both passage and generated questions into the index, and matching incoming questions to stored questions during search, basically an implementation of the doc2query (and subsequently doctT5query) papers. Here are my slides for those interested.

[manisnesan]

Edit : Incorrect tag fixed.

Check out the Lightning Talks from Haystack US 2023 https://youtu.be/Ddss4r1fScc featuring:

Connor Shorten: Filtered Vector Search David Tippett: Using search relevance with Observability Doug Turnbull: Search relevance community supporting each others careers during hard times Peter Dixon-Moses: Flying Blind - Shared inspiration for search analytics Gus Heck : Relevance is super cool, but to do that you need to get documents into your search engine first. That's what JesterJ does Michael Froh: Customizing search requests and responses on OpenSearch with search pipelines Aruna Govindaraju: User signals to track and analyze Uri Goren: Explainable recommendation systems with vector search Brian Nauheimer: Measuring performance for vector search Eric Pugh: Help Atita Measure Vector Search Relevance Accuracy With a Rating Party

Youtube

[manisnesan]

Haystack conference experience

https://www.luminis.eu/blog/introduction-of-frank-and-amy-at-haystackconf-usa/

Trip report by Sujitpal

https://sujitpal.blogspot.com/2023/04/haystack-us-2023-trip-report.html

[manisnesan]

Learning to hybrid search: combining BM25, neural embeddings and customer behavior into an ultimate ranking ensemble was a presentation by the author of Metarank Roman Grebenikkov. He makes the point that lexical (BM25) search is good at a few things and neural search is good at a few other things. Therefore combining the two (or more) searches as an ensemble can address the weaknesses of both systems and improve results. Metarank was used to evaluate this idea using various ensembles of techniques.

Amazon’s ESCI dataset and compare traditional text matching and LTR approaches with neural search methods on real data. We will show how combining multiple old, and new approaches in a in a single hybrid system can deliver an even better result than each of them separately.

[gus]

Check out the Lightning Talks from Haystack US 2023 https://youtu.be/Ddss4r1fScc featuring: ...

I think you tagged the wrong accounts on almost all of these, including my own.

[manisnesan]

Sorry about the wrong tag.