Is Your Infrastructure Ready For Learning To Rank?
The success of a machine learning project will be dependent on the quality and quantity of the data available.
Optimizing User-Product Matching Marketplaces
We explore the parallels between search, recommendation systems and microeconomics. Proposing a new way to think of search relevance and recsys as a system of maximizing the utility of multiple buyers. Is search and discovery akin to quantitative finance in its potential for optimization?
Heatmaps in Elasticsearch
Heatmaps over geo_shape fields in Elasticsearch
Using Splainer to Debug Complex Solr Function Queries
Solr function queries look like Excel queries from hell. I want to tell you how I debug complex Solr function queries using Splainer.
event
OpenSource Connections is proud to host beSwarm 8
OpenSource Connections is proud to again host beSwarm! beSwarm is a gathering of tech enthusiasts from all over central Virginia to discuss, share, and learn about anything that interests us. The idea of the event is that participants “vote with their feet,” moving around the room and forming swarms around interesting discussions and presentations.
When to choose open source search (and when you shouldn't)
In this article, I want to single out Solr and Elasticsearch specifically for consideration. Why? As open source solutions, they share *most* of the pros/cons compared to proprietary solutions when making a business decision.
event
Elastic DC Meetup: Kibana Basics
New to Kibana, and related tools? In April we'll be doing a beginners night for folks interested in Kibana and adjacent topics. We'll have a 25 minute share plus YOUR lightning talks!
event
DrupalCon: Baltimore
Can Drupal+Solr Yield Google-like results? Come chat with CEO Eric Pugh at DrupalCon!
event
RVA Data Hackers
Doug will highlight the Elasticsearch Learning to Rank plugin OSC has developed. Jason Kowalewski will discuss use cases related to personalized job search & recs.
Test Driving Elasticsearch Learning to Rank with a Linear Model
This blog post let's the simple model we derived in a previous post and make it usable via the Elasticsearch Learning to Rank plugin.
Learning to Rank 101 -- Linear Models
Many search nerds get an instinct they want to 'learn the right boosts' to apply to their queries. This blog post covers how to do that using a linear learning to rank model
event
Elastic DC Meetup
Doug will highlight the Elasticsearch Learning to Rank plugin OSC has developed. Jason Kowalewski will discuss use cases related to personalized job search & recs.
event
Future of Data Philly Meetup
Doug will be speaking about 'Elasticsearch as a General Purpose Recommendation System' including highlighting Elastic Graph's capabilities to support a real-time recommendation system.
podcast
Trey Grainger Chats about Solr and Lucene Revolution Part. 2
Trey talks about what is new and upcoming in Solr
What is Learning To Rank?
We walk through Learning to Rank, which uses machine learning to improve the relevance of search results.
event
Elastic{On}
Doug will be attending ElasticOn and sharing OSC's new Elasticsearch Learning to Rank Plugin with the conference attendees.
Solr UTF-8 Character Handling
This article investigates Solr support for non-ASCII utf-8 characters.
We're Bringing Learning to Rank to Elasticsearch
Announcing the Elasticsearch Learning to Rank Plugin: Bringing machine learning into Elasticsearch to improve search relevance.
The Conferences We Went to in 2016
During our Annual Retreat, we looked back at the conferences that we spoke at. I wanted to share ;-)
podcast
Trey Grainger Chats about Solr and Lucene Revolution Part. 1
Trey is SVP of Engineering at Lucidworks and the co-author of Solr in Action
Our Solution to Solr Multiterm Synonyms: The Match Query Parser
You have probably heard us talk about Solr Multiterm Synonyms a lot. This article demonstrates a technique that solves the problem, with fewer performance and relevance issues than other approaches we've used.
OpenBadges: Currently dragging me through the trough of disillusionment
I try not to get sucked in by hype, but OpenBadges managed to do just that, and has left me in the trough of disillusionment. However, maybe a light at the end of the tunnel?
Patterns for Elasticsearch Synonyms: Taxonomies and Managed Vocabularies
How to build semantic search using curated taxonomies and managed vocabularies
event
DC Data Wranglers - It’s a Balloon! A Blimp! No, a Dirigible! Apache Zeppelin: Query Solr via Spark
Eric will be sharing some experiences making Solr part of a larger data ecosystem through using Apache Spark and Apache Zeppelin.
Patterns for Synonyms in Elasticsearch: Keyphrases
In this article I want to begin to discuss patterns that I've found useful when managing Solr/Elasticsearch synonyms.
Traditional Retail: Stop Whining, Start Reverse Showrooming
Today I want to talk about one strategy traditional retailers can use: reverse showrooming. What do I mean by reverse showrooming? Basically that retailers take advantage of my materialistic lust to have an item *right fricken now* while I'm sitting at home.
First Impressions of Apache NiFi
Moving data around. It's always a pain. Is NiFi the better mousetrap for this?
BM25F in Lucene with BlendedTermQuery
As part of the London hack days Diego Ceccarelli started a BM25F implementation. I began to continue it at Lucene Revolution's Lucene hackathon. I realized though that when you break down the problem, BM25F can be implemented using existing Lucene bits, including the existing BM25Similarity and the BlendedTermQuery.
Building Recommendation Systems with Elastic Graph
Graph + Search means easy to implement recommendations. Check out what we've done with the Elasticsearch graph plugin.
event
A One/Two punch of speakers on Testing and Relevancy
Doug Turnbull (OpenSource Connections), ‘Test Driven Relevancy' and Rene Kriegler discusses manages relevance with Querqy
podcast
Search as a Recommendor System
What's new in ES and Solr; Doug goes into more detail on using search engines as recommendor systems.
Expanding data from a frequency table into case form in R and STATA
Sometimes you need to convert from case tables to frequency tables. Here is how I did it with R.
event
beCamp!
If you're a geek in or around the Charlottesville metroplex or even if you're merely tech-curious, this is the event you don't want to miss. beCamp is Charlottesville's version of the BarCamp unconference phenomenon: organized on the fly by attendees, for attendees.
Why I think search engines are the future of recommendation systems
There's a trend away from recommendation systems being the domain of wealthy organizations. Increasingly, I see a shift towards search engines like Solr or Elasticsearch for implementing recommendation systems
High-Quality Recommendation Systems with Elasticsearch
Let's explore how to deliver great recommendations with Elasticsearch. In this article, we dive into an aggregrations based method for Elasticsearch recommendations. We attempt to understand the mechanics and assumptions of the underlying JLH scoring method.
event
Cassandra Summit
Eric will be giving a short talk on how to break the Cassandra data modeling strait jacket with DSE Search
Your Recommendation Systems Aren't As Cool As My Friends
Users listen to their friends, not your recommendation system. Is there a way to reimagine recommendation systems around social influence?
We empower great search teams to achieve their potential
We teach some of the largest organizations how to deliver more accurate, relevant results with Solr and Elasticsearch. If search is core to your value, you should never outsource it.