This repository has been archived by the owner on Apr 22, 2020. It is now read-only.
Releases: neo4j-contrib/neo4j-graph-algorithms
Releases · neo4j-contrib/neo4j-graph-algorithms
3.5.4.0
3.4.12.7
What happened in this release?
- Fixed PageRank overflow bug when working with nodes with very large degrees
- Adapted concurrency in community edition to desktop and entry level server capabilities.
3.5.3.4
What happened in this release?
- Bug fix for Degree Centrality
- Added InfoMap, BFS, DFS
- Renamed
algo.getNodeByIdtoalgo.asNode - Added
algo.asPath - Support for huge graphs for LPA
- UnionFind deadlock bug fix
- Fixed bug with default weight property on shortest path procedure
3.4.12.6
What happened in this release?
- Bug fix for Degree Centrality
- Added InfoMap, BFS, DFS
- Renamed
algo.getNodeByIdtoalgo.asNode - Added
algo.asPath - Support for huge graphs for LPA
- UnionFind deadlock bug fix
- Fixed bug with default weight property on shortest path procedure
3.5.3.3
What happened in this release?
- Added Eigenvector Centrality
- Refactored Strongly Connected Components, Connected Components, Label Propagation to use
writePropertyinstead ofpartitionProperty. Old parameter still works for backwards compatability - Refactored
algo.lpato match signatures of other write algorithms. Old signature still works for backwards compatability
3.4.12.5
What happened in this release?
- Added Eigenvector Centrality
- Refactored Strongly Connected Components, Connected Components, Label Propagation to use
writePropertyinstead ofpartitionProperty. Old parameter still works for backwards compatability - Refactored
algo.lpato match signatures of other write algorithms. Old signature still works for backwards compatability
3.5.3.2
What happened in this release?
- Louvain - support for Random Neighbor Community Selection
- Added Degree Centrality
- Similarity Algorithms now support optional
sourceIdsandtargetIds - Added duplicate neighbors strategy for Cypher loading - can now choose to pick first weight, maximum weight, minimum weight, or sum relationship weights
- Bug fix for k shortest path max depth - thanks to @frant-hartm
algo.list()now returns functions as well as procedures
3.4.12.4
What happened in this release?
- Louvain - support for Random Neighbor Community Selection
- Added Degree Centrality
- Similarity Algorithms now support optional
sourceIdsandtargetIds - Added duplicate neighbors strategy for Cypher loading - can now choose to pick first weight, maximum weight, minimum weight, or sum relationship weights
- Bug fix for k shortest path max depth - thanks to @frant-hartm
algo.list()now returns functions as well as procedures
3.5.3.1
What happened in this release?
-
Link prediction functions - Total Neighbors, Same Community, Preferential Attachment
-
Bug fix for neighbor computation
3.4.12.3
What happened in this release?
-
Link prediction functions - Total Neighbors, Same Community, Preferential Attachment
-
Bug fix for neighbor computation