Largely, it remains, the neglected stepchild of e-commerce optimization. Site-search optimization has the potential to catapult your customer journey strategy to a new level.
This is a practical guide for engineers and product managers about how to combine multiple definitions of item quality to form a “pretty good” overall score of quality using a simple linear model. This isn’t the best or optimal way to optimize user experience, but it’s easy to implement, understand, extend, is generally applicable to virtually any product, and is time-tested in industry.
Unsupervised Attribute Extraction for Online Listings
I will talk about my project on developing an unsupervised approach to extract attributes from online listings, done in collaboration with OLX Group, part of Prosus. The OLX Group operates a network of online trading platforms in over 40 countries, building market leading classifieds marketplaces that empower millions of people to buy, sell, and create prosperity in local communities.
NLP: All the Features. Every Feature That Can Be Extracted From the Text
I will be sharing all the possible NLP features that you can extract from unstructured texts for using in downstream tasks. I also list the python libraries I prefer to use for computing these features.
Query Understanding: An efficient way how to deal with long tail queries
Our data shows that when people search for a certain product, most of them use roughly 1.5 words. These short queries unfortunately make it hard for full-text search to offer them relevant results. While there is improvement to be found in using filters, there are often so many that it can be confusing. One of the ways to make searching more effective is to use the ‘learning to rank’ approach, which creates an optimal ranking of results. However, even this machine-learning method is not all-mighty – and that’s why we’ve come up with Query Understanding, a great companion to ‘learning to rank’.
Despite the fact that site search often receives the most traffic, it’s also the place where the user experience designer bears the least influence. Few tools exist to appraise the quality of the search experience, much less strategize ways to improve it. When it comes to site search, user experience designers are often sidelined like the single person at an old flame’s wedding: Everything seems to be moving along without you, and if you slipped out halfway through, chances are no one would notice. But relevancy testing and precision testing offer hope.
BERT (Bidirectional Encoder Representations from Transformers) turned 2 years a few days ago, and since its introduction it has been a revolution for Search and Information Retrieval. It has drastically improved the accuracy on many different information seeking tasks, be it answering questions or ranking documents, far beyond what was thought possible just a few years ago. In this blog post I’ll give an quick overview of how to evaluate search ranking models using well established relevancy datasets and how to achieve terrible ranking results using BERT in a way it was not meant to be used with a few good pointers on how to successfully apply BERT for ranking.
why the smartSuggest module might matter to you
Guide the user during the search formulation process to facilitate accurate data entry, encourage exploratory search and boost product discovery.
https://blog.searchhub.io/why-weve-developed-the-searchhub-smartsuggest-module-and-why-it-might-matter-to-you
Query Segmentation and Spelling Correction
https://towardsdatascience.com/query-segmentation-and-spelling-correction-483173008981
ELMo Embedding — The Entire Intent of a Query
https://medium.com/analytics-vidhya/elmo-embedding-the-entire-intent-of-a-query-530b268c4cd
Search to Search recommendations (Collaborative Synonym and Spell corrections)
https://haystackconf.com/europe2019/search-to-search-recommendations/
What is Search in the Omnichannel?
https://opensourceconnections.com/blog/2020/12/18/what-is-search-in-the-omnichannel/
Using approximate nearest neighbor search in real world applications
https://blog.vespa.ai/using-approximate-nearest-neighbor-search-in-real-world-applications/
GPT-3: Demos, Use-cases, Implications
https://towardsdatascience.com/gpt-3-demos-use-cases-implications-77f86e540dc1
Roles in a Data Team
Different roles in a data team and their responsibilities
https://towardsdatascience.com/roles-in-a-data-team-d97a87fdabaa
Search Product Management: The Most Misunderstood Role in Search?
https://jamesrubinstein.medium.com/search-product-management-the-most-misunderstood-role-in-search-2b7569058638
Improving search relevance with data-driven query optimization
https://www.elastic.co/blog/improving-search-relevance-with-data-driven-query-optimization
Using Behavioral Data to Improve Search
https://tech.ebayinc.com/engineering/using-behavioral-data-to-improve-search/
Search Product Manager: Software PM vs. Enterprise PM or What does that * PM do?
https://www2.slideshare.net/jt_kane/search-product-manager-software-pm-vs-enterprise-pm-or-what-does-that-pm-do
Analyzing online search relevance metrics with the Elastic Stack
https://www.elastic.co/blog/analyzing-online-search-relevance-metrics-with-the-elastic-stack
Introducing txtai, an AI-powered search engine built on Transformers
Add Natural Language Understanding to any application
https://towardsdatascience.com/introducing-txtai-an-ai-powered-search-engine-built-on-transformers-37674be252ec
Best Practices for Enterprise Search User Experience (UX)
https://www.searchblox.com/best-practices-for-enterprise-search-user-experience-ux/
The Annual Search Shootout – Testing strategy on 2019’s topics
https://opensourceconnections.com/blog/2020/11/25/the-annual-search-shootout-testing-strategy-on-2019s-topics/
Three Pillars of Search Relevancy. Part 1: Findability
https://blog.searchhub.io/three-pillars-of-search-quality-in-ecommerce-part-1-findability
Use Site Search to Optimize Your Customer Journey
Largely, it remains, the neglected stepchild of e-commerce optimization. Site-search optimization has the potential to catapult your customer journey strategy to a new level.
https://blog.searchhub.io/why-use-site-search-analytics-to-optimize-your-customer-journey
Weighted Quality Score for Ads, Feed, and Search
This is a practical guide for engineers and product managers about how to combine multiple definitions of item quality to form a “pretty good” overall score of quality using a simple linear model. This isn’t the best or optimal way to optimize user experience, but it’s easy to implement, understand, extend, is generally applicable to virtually any product, and is time-tested in industry.
https://medium.com/promoted/weighted-quality-score-for-ads-feed-and-search-2fa70ec4f51f
Unsupervised Attribute Extraction for Online Listings
I will talk about my project on developing an unsupervised approach to extract attributes from online listings, done in collaboration with OLX Group, part of Prosus. The OLX Group operates a network of online trading platforms in over 40 countries, building market leading classifieds marketplaces that empower millions of people to buy, sell, and create prosperity in local communities.
https://medium.com/prosus-ai-tech-blog/unsupervised-attribute-extraction-for-online-listings-41baa5d2270e
NLP: All the Features. Every Feature That Can Be Extracted From the Text
I will be sharing all the possible NLP features that you can extract from unstructured texts for using in downstream tasks. I also list the python libraries I prefer to use for computing these features.
https://medium.com/swlh/nlp-all-them-features-every-feature-that-can-be-extracted-from-text-7032c0c87dee
Search (Pt 2) — A Semantic Horse Race
Cutting edge NLP vs traditional search
https://towardsdatascience.com/search-pt-2-semantic-horse-race-5128cae7ce8d
Billion-scale semantic similarity search with FAISS+SBERT
Building the prototype for an intelligent search engine
https://towardsdatascience.com/billion-scale-semantic-similarity-search-with-faiss-sbert-c845614962e2
Visualizing 100,000 Amazon Products
Fast sentence embeddings (fse) enables you to compute sentence embeddings for millions of reviews in only a few minutes.
https://towardsdatascience.com/vis-amz-83dea6fcb059
Query Understanding: An efficient way how to deal with long tail queries
Our data shows that when people search for a certain product, most of them use roughly 1.5 words. These short queries unfortunately make it hard for full-text search to offer them relevant results. While there is improvement to be found in using filters, there are often so many that it can be confusing. One of the ways to make searching more effective is to use the ‘learning to rank’ approach, which creates an optimal ranking of results. However, even this machine-learning method is not all-mighty – and that’s why we’ve come up with Query Understanding, a great companion to ‘learning to rank’.
https://www.luigisbox.com/blog/query-understanding/
Search Optimization 101 – How do you fix a broken search?
https://blog.supahands.com/2020/08/04/search-optimization-101-how-do-you-fix-a-broken-search/
Testing Search for Relevancy and Precision
Despite the fact that site search often receives the most traffic, it’s also the place where the user experience designer bears the least influence. Few tools exist to appraise the quality of the search experience, much less strategize ways to improve it. When it comes to site search, user experience designers are often sidelined like the single person at an old flame’s wedding: Everything seems to be moving along without you, and if you slipped out halfway through, chances are no one would notice. But relevancy testing and precision testing offer hope.
https://alistapart.com/article/testing-search-for-relevancy-and-precision/
philosophe.*
Testing Search
https://www.philosophe.com/archived_content/search_topics/search_tests.html
Assumptions About User Search Behavior
https://www.philosophe.com/archived_content/search_topics/user_behavior.html
How not to use BERT for Document Ranking
BERT (Bidirectional Encoder Representations from Transformers) turned 2 years a few days ago, and since its introduction it has been a revolution for Search and Information Retrieval. It has drastically improved the accuracy on many different information seeking tasks, be it answering questions or ranking documents, far beyond what was thought possible just a few years ago. In this blog post I’ll give an quick overview of how to evaluate search ranking models using well established relevancy datasets and how to achieve terrible ranking results using BERT in a way it was not meant to be used with a few good pointers on how to successfully apply BERT for ranking.
https://bergum.medium.com/how-not-to-use-bert-for-search-ranking-4586716428d9
“Avacado” or Avocado?
A simple search query correction heuristic for the resource-constrained
https://tech.instacart.com/avacado-or-avocado-4b4b78dc0698
10-step checklist to build a great search
https://medium.com/videdressing-engineering/10-step-checklist-for-building-a-great-search-1c8373a97a87