The Semaphore - Smartlogic’s Quarterly Newsletter - September 2019

Posted on: September 06, 2019, by: Ann Kelly

CEO Corner

Leveraging Semantics for Intelligence Gathering Applications

Using a knowledge model as a basis for intelligence gathering applications allows you to bring data, context and meaning together into a single domain that can be understood by human operators and queried programmatically.

Let’s look at Colonel Boyd’s OODA theory as a model for processing a Request for Intelligence (RFI). In this example, we’ll only use the first three phases: OBSERVE – obtain all the relevant data points; ORIENTATE – determine the relationships between data points and identify the most significant for decision-making; DECIDE - develop possible plans for resolving the challenges presented by the significant relationships and data points.

Semantics and knowledge models are key to this process, as it is only possible to determine the significance of a data point once its meaning is expressed in terms of the decision to be taken.

There are four types of knowledge models that can be used in the intelligence-gathering process.

Discovery graphs are part of the data acquisition process and can be used to tag textual content or textual metadata (i.e. transcriptions applied to Audio/visual assets). The tags can include names, organizations, and places, etc. as well as entities from one or more of the other graphs. The markup from the discovery graph includes a URI back to the concept, which is represented in the data, allowing relevant data to be identified and grouped together.
Orientation graphs may be developed during data acquisition or as part of the data landscape and represent the key data elements and how they relate to each other. Orientation Graphs can be queried using SPARQL for automated processing and can be linked with other graphs. A metadata hub, providing a SPARQL interface for querying the properties of assets in other systems would also be regarded as an Orientation Graph.
A Proposition graph reflects the opinion of the analyst or analysts. In a full semantic solution, the proposition graph would include semantic links elements in one or more orientation graphs. The key property of a proposition graph is that it links elements/entities using relationships that are as defined in the master ontology.
Underlying all of this is a Master Ontology – which contains the class definitions and relationships that can be used to create proposition graphs. An aspect of the master ontology will also be the definition of the fact that will be extracted through one or more discovery graphs.

The definitions in the Master Ontology provide the framework for understanding and can be used by all actors; human analysts, applications, or cognitive techniques in the intelligence-gathering process.

Semaphore’s capabilities provide value as part of an overall solution that supports investigations or intelligence gathering:

Automatic assignment of metadata based on document content – all content is automatically analyzed and categorized against a knowledge model and marked up with precise and consistent metadata.
Search enhancement - the knowledge model can be used to drive the user search experience and return documents that are about the search term rather than containing a few keywords.
Model curation - Semaphore allows SME’s to collaborate on the creation and evolution of the model. It is also possible to create a distributed model, where project teams are able to maintain their own definitions, which can be made available to the wider organization or kept localized.
Model update - The model can be updated from other systems of record so that the model can be kept consistent as investigations and evidence-gathering evolves; for example, as themes develop or a set of relationships between known entities becomes more important.
Controlled vocabularies - Semaphore will act as the “dictionary” for terms and definitions throughout an agency and its operations; providing a common language across multiple teams and locations.
Entity extraction - Semaphore extracts entities such as names, organizations, groups, etc. based on a linguistic understanding of the text, not on patterns or dictionaries.
Fact extraction - Semaphore can combine categorization through the knowledge model with entity extraction to identify properties and relationships that are defined within text, which can be stored as metadata with the document or in an RDF triple-store.
Inferencing support - Semaphore’s SPARQL end-point allows queries to be run against the model and the data, for additional insight, without requiring the use of traditional inferencing methods.

Best Regards,

Jeremy Bentley - COE and Founder

Transit authority provides a rich and robust user experience with Semantic AI

Driverless cars on city streets, commercial drones delivering packages, and computer-captained ships navigating the high seas – revolutionary changes that promise improvements in mobility, safety, efficiency, and convenience, but do not guarantee them.

The transportation sector is faced with multiple challenges. How the future unfolds depends on which technologies and services consumers and businesses embrace and how policymakers respond.

Transportation authorities must:

Sharply curb greenhouse gas emissions to slow the rate of climate change
Serve a growing population and cope with worsening highway congestion
Maintain and upgrade a massive system of roads, bridges, ports, waterways, airports, and public transit and determine how to pay for those improvements
Adapt to shifts in trade, energy, and funding sources that affect all modes of transportation

The answers to these and other questions are critically important as transportation plays a central role in society and the economy but is frequently taken for granted. To learn how one major metropolitan transit authority leveraged the Semantic AI capabilities of Semaphore to provide targeted and contextualized user search results, download our complimentary case study.

Announcements

Semaphore 5 – coming soon!

Semaphore 5 is on its way! Check our website’s News and blog for updates.

Smartlogic Semaphore Managed Services

You’ve invested time and energy in developing your model and classification strategies, keeping them up to date ensures continued success. Smartlogic’s Semaphore Managed Services (SSMS) offering guarantees your model and metadata are well managed and kept in sync with your business use case.

Our Information Science experts use a documented and well-tested approach to model development and classification to ensure your results are flexible and reflect the changes in your organization and use case over time.

To learn more about Semaphore Managed Services, contact your Smartlogic Account Manager.

Smartlogic and Blue Prism Integration

Smartlogic’s Semantic AI platform Semaphore provides qualified data within automated processes. Semaphore’s ability to classify, extract and deliver insight - with Blue Prism’s connected RPA, lets organizations effectively and efficiently move their digital transformation journey with intelligent end-to-end automated processes. You can find more information on the Smartlogic/Blue Prism classifier on Blue Prism’s Digital Exchange.

Smartlogic brings Semaphore’s SemanticAI to the Automation Anywhere (AA) Platform

Smartlogic’s Semaphore, adds fact extraction, NLP rules-based classification, semantic integration & visualization services to Automation Anywhere’s Digital Workforce. This allows organizations to gain increased insight and knowledge for their business users from their digital processing initiatives. To learn more about the Smartlogic/AA connection, check out their AA bot store.

Upcoming Events

Smartlogic Attending SDL Connect

If you create, translate or deliver content, this event is for you. SDL Connect will explore every step in the content supply chain and provide relevant insights to support the way you do business. Smartlogic’s Philippe Delorme will be at SDL Connect in Burlingame, CA on October 9^th – 10^th connect with him to find out what’s new at Smartlogic and with the Semaphore platform.

Smartlogic is Platinum Sponsor for Taxonomy Bootcamp London

TBC London is the place to go for essential tips, new approaches, the latest taxonomy applications, and to learn how to make the most of cutting-edge technologies and industry-leading software. The Smartlogic team will be on hand to share Semaphore use cases, best practices and lessons learned, across a broad range of industries, with participants at the event. Don’t miss CEO and Founder Jeremy Bentley’s presentation on Tuesday, October 15^th – Semantic AI and Machine Learning.

Scott Henninger to Speak at Graphorum

Graphorum a vendor-neutral, educational conference focused on bringing together emerging disciplines focused on more intelligent information gathering and analysis will be held in Chicago, IL at the Westin Chicago River North from October 14^th – 17th^th. As the popularity of Graph Technology continues to rise, knowledge graphs, graph analytics, graph databases, graphs, and AI are bringing new innovation and new practical applications to the marketplace. If you are attending don’t miss Smartlogic’s Sr. Solution Architect Scott Henninger’s presentation – Knowledge Graphs and Model-Driven Classification on October 16^th at 10:15 am. Register Today!

Smartlogic is Gold Sponsor for KMWorld 2019

“We’re pleased to announce our support for KMWorld again this year. This year’s theme, Knowledge Sharing in the Age of New Technologies, is very exciting as it focuses
on, cultures, people, processes, and the many different types of technologies supporting organizations as they excel in their industries,” Jeremy Bentley, CEO at Smartlogic.

The Smartlogic team will be on hand to share Semaphore use cases, best practices and lessons learned, across a broad range of industries, with participants at the event. There will be a Smartlogic networking event on Tuesday, November 5th at McCormick & Schmicks at 7 pm for customers, partners and event attendees to connect, share ideas and experiences, and meet with the Smartlogic team. Jeremy Bentley - CEO and Founder and James Morris - Senior Information Scientist, will be presenting at Text Analytics Forum:

Jeremy Bentley - Leveraging Semantic AI to Drive & Calibrate Data Quality
Jeremy Bentley – Ask the Expert Panel
James Morris and Jon Stevens, NLP Software Developer - AbbVie - Delivering Better Biomedical Text Analytics Solutions

KMWorld 2019 is co-located with Taxonomy Boot Camp, Enterprise Search & Discovery, Text Analytics Forum, and Office 365 Symposium.

Smartlogic’s User Community Forum

Our next User Community Forum meeting will be held on Wednesday, October 2^nd at 11 am Eastern. If you haven’t signed up yet DO IT TODAY – it’s a great opportunity to meet and connect with like-minded Semaphore users to exchange tips, tricks and best practices with the Semaphore platform and Knowledge Management topics.

Webinars/Recordings

To learn about all upcoming webinars check out our events page at https://www.smartlogic.com/events

If you missed our Machine Learning Myths vs. Reality – How to use Semantic AI to get Trustworthy AI webinar you can get the playback here.

If you missed The Art of the Possible: A Model-driven Approach to Enhanced Information Discovery webinar you can get the playback here.

Semaphore Training

Smartlogic provides training so your team can get the most from the Semaphore platform. Our fall training kicked off with Semaphore Foundation Training on September 9^th. There is still time to register for Semaphore Advanced, System Administration, and our NEW Fact Extraction Framework training courses. To learn more about each course, visit our training
page. To register for courses, email us at training@smartlogic.com or contact your Smartlogic Account Manager.