Archive by Author

Linkurious v1.5: default captions, Neo4j 3.1 support, AllegroGraph and DataStax built-in search, SSO plugins

The new major release of our graph visualization software Linkurious is now available! This release is called Xenodochial Xenoposeidon. It supports Neo4j 3.1 and provides Neo4j -> Elasticsearch continuous indexing for all versions of Neo4j 3. It integrates AllegroGraph and DataStax Enterprise Graph built-in search engines. More Single Sign-On strategies are now available to authenticate users and the SSO system can be extended by plugins. Linkurious v1.5 is available in two flavors: Starter and Enterprise. Releases always apply to both variants.
Continue Reading →

Introducing Ogma, the Javascript library for large-scale graph visualization and interaction

Linkurious announces the release of Ogma, the Javascript library for large-scale graph visualization and interaction.

State of the art visualization libraries for the Web are unable to display graphs with more than 10 000 nodes and edges. With Ogma is it now possible to display more than 100 000 nodes and 100 000 edges. Ogma is compatible with more than 80 browsers and any devices. Companies can now start integrating a reliable and scalable graph visualization component into their applications.

Ogma performance - large-scale graph visualization

Technology

Modular architecture

Ogma makes visualization of large graphs possible by leveraging the full performances of WebGL combined with a modular architecture. The visualization engine is built for WebGL first ; it supports HTML5 Canvas and SVG with the same level of details, though performance are dramatically limited by these rendering technologies.

Dynamic graphs

The library is designed for graphs with living data where nodes, edges, and their properties can change in real time due to user interactions or updates from data sources.

Level-of-details

A key challenge in visualizing large dynamic graphs is to display as little information possible on screen to avoid getting overwhelmed by the amount of data.
The graphics engine of Ogma provides the ability to define visibility thresholds for various visual components. For example, node texts can be displayed if the node is big enough on screen.

Ready to code with

Developers can get started easily using a unified API and an event model carefully designed. They get access to an extensive documentation website that contains the documentation reference of the API, more than a hundred examples, and advanced demos. They usually get answers in less than a single business day through the support channel.

Ogma API documentation

Ogma API documentation

Developers get access to the Ogma Showcase. It is a complete application written in Angular 2 to demonstrate the main features of Ogma. They can load sample graphs into the Ogma Showcase, play with layouts, filters, and styles, and export Javascript code snippets to quickly get results.

Ogma Showcase screenshot

Ogma Showcase screenshot

Integration

Ogma can be integrated into any modern Web-based application. It supports module bundlers such as WebPack, it can be used with React, and provides Typescript definitions for fast integration with Angular 2.

Ogma and its layout algorithms can be run on a server within a Node.js application. It provides a solution for clients running on slow machines so they can delegate computationnally expensive operations to the server.

Features

Ogma comes with a complete set of features, including:

Load your data

  • Call the graph API to add nodes and edges to the visualization.
  • Import data from multiple file formats.
  • Connect to a Neo4j server using the Bolt protocol.
  • Or generate synthetic graphs using Wattz-Strogatz or Barabasi-Albert models.

Customize visual results

  • Apply graph layouts (force-directed, hierarchical, concentric, geospatial).
  • Provide built-in user interactions with the mouse, trackpad, or through the API to hover, to select, to drag, or to group nodes and edges.
  • Customize style of nodes and edges with both basic visual variables (color, size, shape, icon, image) and advanced visual components (badges, stroke, outline, halo, and pulse effects).
  • Customize style of texts of nodes and edges (font, color, position, style, size, background) with support for line wrapping and line breaks.

ogma-example-customization

Help users analyze graph data

  • Create data-driven styles: map properties of nodes and edges to visual variables such as colors, sizes, and icons for easy and always up-to-date customization.
  • Filter nodes and edges, with the ability to create compound filters with AND/OR operators.
  • Group nodes and edges manually or with specific rules.
  • Find shortest paths.
  • Display a legend to communicate easily the meaning of the visual variables.
  • Display tooltips on user interaction to provide contextual information on nodes and edges, or to implement a context menu.

Export your results

  • Export customizable image files to SVG, PNG, JPG, TIFF which can integrate a custom watermark.
  • Export data in multiple file formats.

Roadmap

We have focused our efforts on building the next generation of interactive graph visualization engine for Web-based applications. We are now improving the scalability and quality of existing layouts. We will provide you more information during the coming months.

Deprecation of linkurious.js

Ogma comes with all the features of linkurious.js, which becomes deprecated. We will continue to support existing customers of linkurious.js until they migrate to Ogma.

How to get it

Ogma is available in a proprietary license only. Contact us to evaluate it!

Meet Francesco Infante, back-end engineer @ Linkurious

The Linkurious team is growing and we are glad to welcome a new member. Francesco Infante is joining us as Back-end Engineer. Welcome Francesco!

infante-portrait

Who are you?

I’m a software engineer, but that’s just the tip of the iceberg. I’ve always knew what my passion was; I’ve started at the age of 3, playing Commander Keen 4 and Wolfenstein 3D over and over, and learning MS-DOS enough to boot those games. Since then my passion never changed; while other kids were playing football, I was learning Visual Basic. My best Christmas memory was when my dad brought home my first pair of PC audio speakers! After that there was no going back :-).

Can you tell us about the products you love?

I love when a product is intuitive enough that I don’t have to learn how to use it. If I can deduce how it works by only glancing at it, it’s a good product for me. If somehow it’s too complicated to understand, it’s the designer fault. Since my expectations are high in this matter, I plan to make my work, in particular in the Linkurious back-end, as intuitive as possible within my ability; both for team-mates and for the final user.

What do you expect to learn at Linkurious?

I want to boost my analytical skills. Gladly, I work with bright people; there’s nothing worse than being the smartest in the room. I hopefully will continue facing new problems and solving them for the most part; learning from the unresolved ones too. I’m also looking into improving my communication and leading skills; and learning French of course!

Which super hero or fictional character would you be?

I would be the Doctor of course! All that knowledge of space and time in my mind… may probably be too much; but for a chance to see what the universe hides, I would bear the burden :-).

What is one thing few people know about you?

I like to pick locks as a hobby; still an enthusiast, though.

What is your favourite music piece at the moment?

Linkurious v1.4: streamline pattern detection with the Cypher-powered alert dashboard

The new major release of our graph visualization software Linkurious is now available! It provides a unique alert dashboard for teams of analysts, continuous indexing capabilities to Neo4j, support for AllegroGraph, and single sign-on (SSO) with Azure Active Directory. Linkurious v1.4.2 is available in two flavors: Starter and Enterprise. Releases always apply to both variants.

New features

Cypher-powered Alert Dashboard (Enterprise only)

From anti-money laundering and insurance fraud detection to cyber-security, companies must identify threats in connected data. The Neo4j graph database is perfectly suited to run complex queries and to extract patterns of suspicious or abnormal activity with the Cypher query language. Linkurious 1.4 provides a solution to streamline threat detection with its Alert Dashboard. Teams of analysts can now combine the power of graph-based analytics with human intelligence to detect and confirm alerts collaboratively.

Example of alert categorization and audit trail.

Alert categorization example.

It works as follows:

  1. Administrators create alerts with the Cypher query language.
  2. Alerts are generated at a given frequency.
  3. A team of analysts review the alerts to confirm or dismiss them. Analysts can sort alerts based on their importance, for instance money or risk involved. They can see who is working on alerts to synchronize with each other, and they can get the audit trail of alerts.

Learn more on the user manual and admin manual.

Example of an user alert dashboard

Introducing the user alert dashboard.

Notice that it is also possible to use the Alert Dashboard with an external pattern detection system instead of a graph database. All features can be easily integrated into third-party systems thanks to our REST API.

allegrograph-logo

Connector to AllegroGraph

AllegroGraph is industry leader of semantic database systems. We have teamed up with Franz, Inc to support AllegroGraph in Linkurious. Linkurious is now a web-based alternative for Gruff users to explore their semantic databases. They can benefit from all features of Linkurious, including data editing, text configuration, and raw SPARQL queries. Contact us to evaluate it!

Screenshot of Linkurious with DBpedia on Franz AllegroGraph

Neo4j continuous indexing

Linkurious will integrate seamlessly with Neo4j graph databases of unlimited size with the continuous indexing system. It is now possible to synchronize Neo4j 3.0.4+ databases with the Elasticsearch index automatically, without blocking users. We have contributed to neo4j-to-elasticsearch, the open source Neo4j plugin initially created by our friends at GraphAware. Kudos to them! Notice that we currently required an initial indexing phase to bootstrap Linkurious.

Single Sign-On with Azure Active Directory (Enterprise only)

You can configure Linkurious for a seamless authentication experience and secure access management with Azure Active Directory.

Linkurious login page with Azure AD

Noteworthy improvements

Support of Elasticsearch 2

Linkurious 1.4 is compatible with Elasticsearch 2.x. We continue to ship Linkurious with Elasticsearch 1.4.5 until considered totally safe to make the switch. You can now configure Linkurious to use more recent versions if you need to.

Security improvements (Enterprise only)

  • Users linked to an LDAP account can be removed.
  • Allow only users who belong to a configurable list of LDAP groups.
  • Assign a configurable user group for LDAP users the first time they connect to Linkurious.
  • Neo4j credentials can be read from environment variables instead of being stored in the configuration file of Linkurious.
  • Read-only mode can be independently enabled for each datasource.

Other improvements

  • Administrators can reset styles of newly created visualizations of all users from the Data administration panel.
  • You can create a widget from a visualization ID.
  • You can apply a force-directed layout computed on the server when creating a new visualization.
  • You can disable indexation of edges.
  • The “shortestPath” feature in Cypher queries is now allowed.
  • Node and edge count are more robust in DataStax Enterprise Graph.
  • Retry on error in DataStax Enterprise Graph indexation.
  • Fix truncated widget content when MySQL is used as internal data store.

What’s next?

The next release will integrate DataStax Enterprise and AllegroGraph full-text search engines for increased performance. Elasticsearch will become optional. We will also support more authentication providers based on your requests.

Try now

Try this new release on our online demo. We have just updated it! Contact us for any question, or send an email to support@linkurio.us for help.

Existing customers can download the new release from their account now.

Linkurious v1.3.6: DataStax Graph 5.0 (final), UI improvements

The new minor release of our graph visualization software Linkurious is now available! It provides the final version of the connector to DataStax Enterprise Graph 5.0, improves the user interface based on user feedback, and fix multiple bugs. Linkurious v1.3.6 is available in two flavors: Starter and Enterprise. Releases always apply to both variants.

New features

datastax-logo

Datastax Enterprise Graph connector leaves BETA

Linkurious is now compatible with the stable release of Datastax Enterprise Graph. See a live demo here. Linkurious connects to it through the Gremlin server. More info here.

This is a sample configuration of Linkurious to connect to DataStax Enterprise Graph 5.0 (DSE) through a Gremlin server.

"dataSources": [
  {
    "name": "My DSE DB",
    "graphdb": {
      "vendor": "dse",
      "url": "ws://192.168.0.45:8282", // URL of the gremlin server to connect to
      "graphName": "mygraph", // name of the graph to connect to
      "create": false // whether to create graphName if it does not exist
    },
    "index": {
      "vendor": "elasticSearch",
      "host": "127.0.0.1",
      "port": 9201,
      "forceReindex": false,
      "dynamicMapping": false
    }
  }
]

incremental-layout

Noteworthy improvements

Incremental layout at expand

Users can enable the “incremental expand” mode for layouts. Existing nodes will hold their positions while the neighbors are added to the visualization. Use it to create hand-made layouts safely.

Letters and numbers as node icons

text-icon
The Design panel of the Workspace now provides the ability to display numbers and latin letters inside nodes.

Select shortest path extremities from the database

path-autocomplete
We have improved the shortest path interface with the ability to select nodes from the database directly using the search autocomplete feature.

Welcome-in-Arabic-Script-300x145

Optimized multilingual search

Integrators can configure Linkurious to use a custom ElasticSearch analyzers (list here). It will enhance search results on a specific language. Try it on Arabic!

Other improvements

  • Add “Create new visualization” in Workspace menu.
  • Rename “Duplicate to Workspace” to “Save As …” in Workspace menu.
  • “Back to dashboard” from Workspace goes to current visualization’s folder.
  • Add “Hide unselected items” in Workspace menu.
  • Add a configuration option to disable new LDAP user creation.
  • URI detection: support schemes such as “notes://”.

What’s next?

We are very excited by the next release as it will bring major new features. Teams of analysts will benefit from the ability to automatically identify graph patterns via a new alert dashboard. Linkurious will integrate seamlessly with graph databases of unlimited size with the continuous indexing system. We expand the reach of our graph platform with the support of the industry-acclaimed semantic database Franz AllegroGraph, soon following a generic SPARQL connector. We will also provide finer authentication control with LDAP groups, and server-side creation of visualization widgets through the REST API.

Try now

Try this new release on our online demo. We have just updated it! Contact us for any question, or send an email to support@linkurio.us for help.

Existing customers can download the new release from their account now.

Meet Maxime Allex, front-end engineer @ Linkurious

The Linkurious team is growing and we are glad to welcome a new member. Maxime Allex is joining us as Front-end Engineer. Welcome Maxime!

photo

Who are you?

I’m a front-end developer dad, formed in art school. Amateur of good wine and cold beer, I also drink water and coffee because I am curious by nature. I am passionate about my children, web, interfaces and books.

Can you tell us about the products you love?

I’m obsessed by the « naturalness » of things, possibly because of my formation in graphic design. I worked a lot on the book object, and I’m still amazed by the simplicity and the naturalness of the experience it procures. You learn to use a book and its purpose before learning to speak or walk. And this simple object can tell you stories on a lot of different levels. Despite the fact that its content is the more concret or factual you can find, the feelings it procures are very different for each person. From my point of view, graph visualization is quite close. You can make graphs tell everything you want. It’s open to all possible interpretations from the editor to the viewer, based on pure data and in a strictly graphical way and that’s really cool.

To getting back to the question, I use a lot of great web products everyday. But in terms of experience, I’m never completely satisfied. There’s always a phase of adoption more or less painful, and in the end, it’s never fully « natural ». However, this is precisely what I love about the web. We are at a time where software experience need / try to be reinvented. We leave the paradigm of windows to the canvas of a web page or a phone, with new interactions that the medium induced. This is a world of possibilities and I would like to contribute.

What do you expect to learn at Linkurious?

I hope to improve my coding skills in contact of talented colleagues, to learn and deepen topics related to the visualization of graphs and its impact. I also wish to help Linkurious in its development, see the company grow, which will allow me to meet even more talented colleagues.

Which super hero or fictional character would you be?

I would be Aragorn Skywalker.

What is one thing few people know about you?

I have 2 kids, 3 cats and I farm vegetables in Paris’ suburb.

What is your favourite music piece at the moment?


and many many others.