Fusion AI 4.0.0 Release Notes
Release date: 21 February 2018
Fusion AI is a license-controlled subset of Fusion features. Starting with Fusion 4.0, a Fusion Server license enables the basic Fusion feature set, while a Fusion AI license enables these additional features:
-
Signals
-
App Insights (new in 4.0)
-
Smart jobs
-
ALS Recommender Jobs
-
Bisecting KMeans Clustering Jobs
-
Cluster Labeling Jobs
-
Co-Occurrence Similarity Jobs
-
Collection Analysis Jobs
-
Document Clustering Jobs
-
Ground Truth Jobs
-
Head/Tail Analysis Jobs (new in 4.0)
-
Item Similarity Recommender Jobs
-
Levenshtein Spell Checking Jobs
-
Logistic Regression Classifier Training Jobs
-
Matrix Decomposition-Based Query-Query Similarity Jobs
-
Outlier Detection Jobs
-
Random Forest Classifier Training Jobs
-
Ranking Metrics Jobs
-
SQL-Based Experiment Metric Jobs
-
Statistically Interesting Phrases Jobs
-
Token and Phrase Spell Correction Jobs (new in 4.0)
-
-
REST APIs
-
Experiments API
-
Recommendations API
-
Signals Aggregator API
-
Signals API
-
-
Index pipeline stages
-
Machine Learning index stage
-
OpenNLP NER Extractor index stage
-
Signal Formatter index stage
-
Update Experiment index stage
-
-
Query pipeline stages
-
Advanced Boosting query stage
-
Analytics Catalog query stage
-
Experiment query stage
-
Machine Learning query stage
-
More Like This query stage
-
Recommend Items for Item query stage
-
Recommend Items for User query stage
-
Recommendation Boosting query stage
-
See also the Fusion Server 4.0.0 release notes.
New features
-
App Insights
App Insights provides detailed, real-time, searchable reports and visualizations derived from your signals data. It also provides alerts and triggers to notify you when specific events occur.
Pre-4.0 signals data will not produce useful visualization in App Insights. New signals generated with Fusion 4.0 will produce the best results.
-
New jobs
-
Perform head/tail analysis of queries from collections of raw or aggregated signals, to identify underperforming queries and the reasons. This information is valuable for configuring better synonyms, auto-suggest, recommendations, and so on, in order to improve conversion rates.
-
Token and Phrase Spell Correction
Extract tail tokens (one word) and phrases (two words) and find similarly-spelled head tokens and phrases. If several matching heads are found for each tail, the job can compare and pick the best correction using multiple configurable criteria.
-
Improvements
-
Experiment management features
The Fusion UI now includes interfaces for running and managing experiments.
-
SQL for aggregations
By default, aggregation jobs are now defined using Spark SQL. To use the old aggregation configuration scheme in the Fusion UI, click Advanced in the Aggregation job configuration panel.
Other changes
-
Bisecting Kmeans Clustering Jobs are now called Document Clustering Jobs
-
The data structure for signals has changed
See Signals for complete details about the new structure.
-
The Experiments API is no longer experimental
The API and other experiments features are now fully functional.
-
Aggregator API is deprecated
The
/aggregator
endpoints are deprecated in this release. Aggregations are now managed through the/spark
endpoints.