Shining UI
The user interface guides you into a much more satisfying RRE experience
RRE-Enterprise is built on top of the acclaimed Open Source Rated Ranking Evaluator library to provide our clients with enterprise-level software.
It offers all the goodies from the RRE library and wraps them up with a comprehensive UI and a simplified user experience, that makes Search Quality Evaluation much easier and intuitive.
Click on the image to zoom in
Ratings
Define the ground truth for the Search Quality Evaluation
Ratings define how good the document D is for the query Q as a golden truth. Having reliable ratings is fundamental for the task of Search Quality Evaluation.
RRE-Enterprise supports both Explicit and Implicit ratings: you can supply your explicit ratings in Json to the platform or simply push user interactions (clicks, views, sales…) and RRE-Enterprise will do the hard work for you and estimate the ratings automatically!
Target Search Engine
You have the freedom to spin up an embedded experiment or target your own search engine
RRE-Enterprise supports both possibilities!
Do you have your own search engine external to RRE and ready to use in a QA environment? you can evaluate it directly.
Do you just have the search engine configurations?
Let’s spin up an embedded one and fill it with your test data!
Click on the image to zoom in
Search Technologies supported
Elasticsearch, Apache Solr and Opensearch are supported!
RRE-Enterprise supports the most popular Apache Lucene based Search Engines as targets for the evaluation. Are you using another Search Engine? Let us know and we can give you an estimation for the integration!
Search Engine Configurations
Spinning up an embedded experiment has never been easier and more flexible!
RRE gives you the possibility of setting up experiments that take a search engine configuration dir and spin up an embedded search engine for the evaluation.
RRE-Enterprise steps that up, allowing you to upload the search engine configuration files directly in the UI from your computer or from your GIT repository branches and commits where you diligently keep control of your source.
Click on the image to zoom in
Click on the image to zoom in
Compatible with Search-APIs
Using a custom layer of REST APIs between the front end and your search engine is not a problem anymore!
RRE-Enterprise, through the powerful Query Discovery component, is able to work directly against your search-API to extract the final queries that reach your target search engine.
So you don’t need to spend hours manually defining your query templates anymore!
Metrics
Measure how good your search engine is performing from various angles
RRE-Enterprise supports a vast array of metrics:
-Ndcg@k -ERR -Precision@K -Recall -Average Precision
-F1 -Reciprocal Rank …
Do you need any additional metrics? Let us know!
Search Quality Evaluation Overview
You don't need to be a software engineer anymore to fully appreciate RRE
You prepared your ratings, configured a target search engine, selected a few metrics (Precision, Recall, Ndcg…) and ran an evaluation.
The search quality evaluation dashboard has a handy Overview of the results, showing the historical progress of the metrics of your interest across your data collections and experiments, warnings are raised automatically in case of suspicious results.
Search Quality Evaluation Exploration
if you are a Search Software Engineer there's plenty to debug and analyse!
The dashboard’s Exploration and Compare areas give you advanced control over the granularity of the studies you want to make on your search quality assessment.
You can explore the metrics aggregation over topics, query groups, and queries. It is even possible to directly investigate the results returned by the query to understand why we got a certain score for a certain metric.
READY TO START?
ONE CLICK INSTALLATION
Don’t forget to check our GitHub and follow the instructions to install.
Fill out this form to receive your free product key!
Read the documentation
If you want to discover more about Rated Ranking Evaluator Enterprise, we recommend reading these comprehensive blog posts we wrote about it!