Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

Introduction to Knowledge Graphs: Data Summit 2020

This workshop presentation from Enterprise Knowledge team members Joe Hilger, Founder and COO, and Sara Nash, Technical Analyst, was delivered on June 8, 2020 as part of the Data Summit 2020 virtual conference. The 3-hour workshop provided an interdisciplinary group of participants with a definition of what a knowledge graph is, how it is implemented, and how it can be used to increase the value of your organization’s datas. This slide deck gives an overview of the KM concepts that are necessary for the implementation of knowledge graphs as a foundation for Enterprise Artificial Intelligence (AI). Hilger and Nash also outlined four use cases for knowledge graphs, including recommendation engines and natural language query on structured data.

  • Be the first to comment

Introduction to Knowledge Graphs: Data Summit 2020

  1. 1. INTRODUCTION TO KNOWLEDGE GRAPHS A Presentation for Data Summit Connect 2020 June 8, 2020
  2. 2. INTRODUCTIONS SARA NASH TECHNICAL ANALYST ENTERPRISE KNOWLEDGE JOE HILGER COO AND PRINCIPAL CONSULTANT ENTERPRISE KNOWLEDGE @EKCONSULTING
  3. 3. AGENDA @EKCONSULTING 1 2 3 Foundations of Knowledge Graphs Knowledge Graphs for Artificial Intelligence Use Cases for Knowledge Graphs
  4. 4. WHAT YOU WILL LEARN TODAY @EKCONSULTING How to build a business case for Knowledge Graphs and Enterprise AI The foundations and technical infrastructure to make Knowledge Graphs a reality Practical use cases for Knowledge Graphs: Recommendation Engine, Natural Language Querying, Relationship Discovery, Data Management Where to begin in Knowledge Graph development – developing an ontology
  5. 5. WHAT IS A KNOWLEDGE GRAPH
  6. 6. 90% of the data and information we have today was created just in the past two years. Most organizations are built to organize and manage data and information by type and department or business function. 80% of leaders say their systems don’t talk to each other. Over 85% of the content and information we work with is unstructured. CONFRONTING TODAY’S INFORMATION MANAGEMENT CHALLENGES 90% AI is set to be the key source of transformation, disruption, and competitive advantage in today’s fast changing economy, contributing to 45% of total economic gains by 2030. @EKCONSULTING
  7. 7. FOLKSONOMY Free-text tags. CONTROLLED LIST List of pre-defined terms. Improves consistency. TAXONOMY Pre-defined terms & synonyms. Hierarchical relationships. Improves consistency. Allows for parent/child content relationships. Capture related data. Integration of structured and unstructured information. Linked data Store. Architecture and data models to enable machine learning (ML) and other AI capabilities. Drive efficient and intelligent data and information management solutions. ONTOLOGY Predefined classes & properties. Expanded relationship types. Increased expressiveness. Semantics. Inference. KNOWLEDGE ORGANIZATION CONTINUUM @EKCONSULTING KNOWLEDGE GRAPHS
  8. 8. tax·on·o·my (tāk-sōn-mē) n. pl. tax·on·o·mies 1. The classification of organisms in an ordered system that indicates natural relationships. 2. The science, laws, or principles of classification; systematics. 3. Division into ordered groups or categories: "Scholars have been laboring to develop a taxonomy of young killers" (Aric Press). EK’s Definition of Taxonomy Controlled vocabularies used to describe or characterize explicit concepts of information, for purposes of capture, management, and presentation. BUSINESS TAXONOMY @EKCONSULTING
  9. 9. A defined data model that describes structured and unstructured information through: • entities, • their properties, • and the way they relate to one another. • Ontology is about things, not strings. • Ontologies model your domain in a machine and human understandable format. • Ontologies provide context. • Effective ontologies require a deep understanding of the knowledge domain. BUSINESS ONTOLOGY @EKCONSULTING
  10. 10. § A knowledge graph is a specialized graph or network of the things we want to describe and how they are related § It is a semantic model since we want to capture and generate meaning with the model “The application of graph processing and graph DBMSs will grow at 100 percent annually through 2022 to continuously accelerate data preparation and enable more complex and adaptive data science.” – Gartner’s Top 10 Data and Analytics Technology Trends for 2019 Google’s knowledge graph is a popular use case KNOWLEDGE GRAPH @EKCONSULTING
  11. 11. ONTOLOGY VS. KNOWLEDGE GRAPH @EKCONSULTING
  12. 12. § Consists of triples § concept → relationship → concept § A linked data store that organizes structured and unstructured information through: § entities, § their properties, § and relationships. § Also known as: § Linked Data Store (LD Store) § Triple Store § “Knowledge Graph” Subject Predicate Object Project A hasTitle Title A Person B isPMOn Project A Document C isAbout Topic D Document C isAbout Topic F Person B IsExpertIn Topic D … … … GRAPH DATABASE @EKCONSULTING
  13. 13. Content & Data Sources Subject Predicate Object Person B isPMOn Project A Document C isAbout Topic D Document C isAbout Topic F Person B IsExpertIn Topic D Business Ontology Triple Store/Graph Database Enterprise Knowledge Graph Person B Project A Document C Person F Topic D Topic E Business Taxonomy HOW IT ALL FITS TOGETHER @EKCONSULTING
  14. 14. Resource Description Framework SQL, SAP Structured Data Source Semantic Enterprise Search Analytics Enterprise Knowledge Graph CMS, DMS, CRM, etc. Taxonomy & Auto-tagging Resource Description Framework Unstructured Data Source Question Answering SQL, SAP, Excel Structured Data Source Virtual Mapping ENTERPRISE KNOWLEDGE GRAPH @EKCONSULTING
  15. 15. ARTIFICIAL INTELLIGENCE (AI)
  16. 16. ARTIFICIAL INTELLIGENCE (AI) IN ACTION AI FOR DATA AND INFORMATION MANAGEMENT ENTAILS LEVERAGING MACHINE CAPABILITIES TO DISCOVER AND DELIVER ORGANIZATIONAL KNOWLEDGE AND INFORMATION IN A WAY THAT CLOSELY ALIGNS WITH HOW WE LOOK FOR AND PROCESS INFORMATION. @EKCONSULTING
  17. 17. @EKCONSULTING DECONSTRUCTING AI: DRIVERS BUSINESS AGILITY AGING INFRASTRUCTUREDATA DYNAMISM Volume and dynamism of organizational data/content (structured and unstructured) Growing digitalization, aging of systems and disparate sources User experience, knowledge loss, bad info/data, data team efficiency
  18. 18. DECONSTRUCTING AI: MACHINE LEARNING Inferred Relationships Automatically discover implicit facts in your data Clustering Detect fraud, identify risk factors, categorize customer behavior Auto- Classification Automatically route incoming requests to appropriate channels Machine Learning Image & Shape Recognition Digital Asset Management, product identification, security, intelligence Predictive Analytics Customer retention, risk modeling, predictive maintenance Recommendation Engine Discover new content and information based on context at the point of need Natural Language Processing Simplify user experience, bring data closer to business users @EKCONSULTING
  19. 19. Aggregation, Reasoning, and Optimization Graphs allow for aggregation of information from multiple disparate solutions, which allows users to find information that exists in multiple locations, and optimizes data management and governance. ENTERPRISE KNOWLEDGE GRAPHS & AI Understanding Context Relationships between information give us a better understanding of how things fit together, adding knowledge to data. Structured and Unstructured Information Allows for the organization and integration of structured and unstructured information so that users can search for data and content at the same time. Intuitive Interactions Graphs store information in the way people speak and process information, while simultaneously making it machine readable and therefore ready for human centered applications, such as natural language search. Discover Hidden Facts & Patterns Inferencing allows for large scale analysis and identification of related topics and things. @EKCONSULTING
  20. 20. USE CASES FOR KNOWLEDGE GRAPHS
  21. 21. USE CASE #1: RECOMMENDATION ENGINE
  22. 22. SLIDE WITH CIRCLE PHOTO The Business Challenge A global development bank needed a better way to disseminate information and expertise to all of their staff so that they could complete projects more efficiently, without rework and knowledge loss. Their information and expertise were contained in thousands of unstructured documents and publications that needed to be better organized and made accessible. The Solution ü EK developed a semantic hub, leveraging a knowledge graph that collects organizational content, user context, and project activities. ü The information powered a recommendation engine that suggests relevant articles and information when an email or a calendar invite is sent on a given topic or during searches on that topic, which will eventually power a chatbot as part of a larger AI Strategy. ü These outputs were then published on the bank’s website to help improve knowledge retention and to showcase expertise via Google recognition and search optimization for future reference. Outcomes Using knowledge graphs based on this linked data strategy enabled the bank to connect all of their knowledge assets in a meaningful way to: § Increase the relevancy and personalization of search. § Enable employees to discover content across unstructured content types, such as webinars, classes, or other learning materials based on factors such as location, interest, role, seniority level, etc. § Further facilitate connections between people who share similar interests, expertise, or location. @EKCONSULTING USE CASE #1: RECOMMENDATION ENGINE
  23. 23. @EKCONSULTING USE CASE #1: RECOMMENDATION ENGINE Graph Database
  24. 24. Because of a Knowledge Graph… ü Ability to support future business questions and needs that are currently unknown ü Greater flexibility to quickly modify and improve data flows aligned to business needs ü Flexibility to add new data sources without making extensive changes to data architectures and schemas resulting in rapid iteration and quick adaptation to changing requirements ü Architecture allows to quickly iterate and grow new products and services for its users @EKCONSULTING Recommendation Engine
  25. 25. USE CASE #2: NATURAL LANGUAGE QUERYING ON STRUCTURED DATA
  26. 26. SLIDE WITH CIRCLE PHOTO The Business Challenge One of the largest supply chains needed to provide its business users a way to obtain quick answers based on very large and varied data sets. The data sets were stored in a large RDBMS data warehouse with little to no context, making it difficult to understand its value, which information to use, and what questions it could answer. The goal was to bring meaningful information and facts closer to the business to make funding and investment decisions. The Solution ü By extracting topics, places, people, etc. from a given file, EK developed an ontology to describe the key types of things business users were interested in and how they relate to each other. EK mapped the various data sets to the ontology and leveraged semantic Natural Language Processing (NLP) capabilities to recognize user intent, link concepts, and dynamically generate the data queries that provide the response. Outcomes In doing so, non-technical users were able to uncover the answers to critical business questions such as: § Which of your products or services are most profitable and perform better? § What investments are successful, and when are they successful? § How much of a given product did we deliver in a given timeframe? § Who were my most profitable customers last year? § How can we align products and services with the right experts, locations, delivery method, and timing? @EKCONSULTING USE CASE #2: NATURAL LANGUAGE QUERYING
  27. 27. FVC & LVC Data Virtual Graph Mapping Graph Search Knowledge Graph IDE Configure Graph Mapping Query Graph Data Connects to Graph DB Virtualizes Relational Data Data SME Taxonomy & Ontology Manager SPARQL Knowledge Graph Business User Front End UI Relational NoSQL Metadata External Internal Chatbot Q&A Semantic Enterprise Search NLP @EKCONSULTING USE CASE #2: NATURAL LANGUAGE QUERYING
  28. 28. Because of a Knowledge Graph… @EKCONSULTING ü Rapid alignment of data elements with natural language structure of English questions to identify user intent ü Flexible mapping of disparate data source schemas into a single, unified data model that is “whiteboardable”- accessible to both technical and nontechnical users ü Clear definition of key information entities and their relationships to each other to unleash the value of data contexts and meaning Natural Language Querying on Structured Data
  29. 29. USE CASE #3: RELATIONSHIP DISCOVERY THROUGH UNSTRUCTURED DATA
  30. 30. The Business Challenge A federally funded research and development center (FFRDC) has an extensive “project library” where they store technical documents, certifications, and reports related to various engineering projects. These documents often don’t have much associated metadata and are very difficult to search. When employees start working on new projects, it’s hard to tell, from the project libraries, what was done on previous projects and who did the work. @EKCONSULTING The Solution ü Using an existing business taxonomy developed by the FFRDC, EK led the development of an enterprise knowledge graph, connecting documents to projects, topics, and individuals through auto-tagging ü EK also developed a semantic search platform, enabling document searches based on context. Outcome Using the enterprise knowledge graph, the FFRDC could then use the semantic search application to § Browse documents by person, project, and topic § Analyze relationships between people and projects directly USE CASE #3: RELATIONSHIP DISCOVERY THROUGH UNSTRUCTURED DATA
  31. 31. ▪ Enhanced Auto-Tagging ▪ History of Documents ▪ Implicit Auto-Tagging ▪ Associate Taxonomy Terms ▪ Classification ▪ Group Content based on Tags Taxonomy Content Tag Co-occurrence @EKCONSULTING USE CASE #3: RELATIONSHIP DISCOVERY THROUGH UNSTRUCTURED DATA
  32. 32. v PROJECTS PEOPLE TOPICS showing 53 results for PROJECT X... Project X John Doe (25) Emily Smith (14) Robert Jones (5) Topic A (19) Topic B (11) Topic C (3) Related People Related Topics @EKCONSULTING USE CASE #3: RELATIONSHIP DISCOVERY THROUGH UNSTRUCTURED DATA
  33. 33. Because of a Knowledge Graph… @EKCONSULTING ü Ability to support future business questions and needs that are currently unknown ü Greater flexibility to quickly modify and improve data flows aligned to business needs ü Flexibility to add new data sources without making extensive changes to data architectures and schemas ü Architecture allows to quickly iterate and grow new products and services for its users RELATIONSHIP DISCOVERY THROUGH UNSTRUCTURED DATA
  34. 34. USE CASE #4: DATA MANAGEMENT
  35. 35. The Business Challenge The data scientists and economists at the Federal Agency were having trouble connecting siloed data sources to easily access, interpret and track all the data and history in order to provide meaningful context to the Board. This Agency needed a solution that enhanced and modernized their metadata management practices through improved access and visibility across their data resources while maintaining the appropriate security. @EKCONSULTING Solution ü EK led the development of an advanced, semantic metadata modeling prototype, leveraging a knowledge graph to provide key contextual and descriptive information that helped map relationships across the Agency’s regulatory data sources. ü EK also developed an intuitive front-end user interface that enabled end-users and data SMEs to explore and access the data in the model. The model made it easy to find and connect to the relevant data the business user needs to view key information at a glance. Outcome Data analysts and researchers can now: § Access to the Agency’s data resources in a single tool that makes data stored in multiple locations available without moving or copying the data. § Spend less time tracking or processing data for non-technical users who can now directly access and explore the data for decision making. USE CASE #4: DATA MANAGEMENT
  36. 36. Because of a Knowledge Graph… @EKCONSULTING ü Achieve powerful alignment between the application UI and knowledge graph structure allowing the graph to define the templates that the UI populated with key data from the graph ü Encourage the users to explore the information by traversing relationships that made navigating the data easy and intuitive ü Arrange the information from both unstructured documentation and structured data sources into a single, structured format ü Optimize data quality by allowing the analysis of network effects, through patterns DATA MANAGEMENT
  37. 37. WE’LL BE ANSWERING QUESTIONS NOW Q A& THANKS FOR LISTENING Q & A SESSION
  38. 38. CONTACT US WWW.LINKEDIN.COM/IN/JOSEPH-HILGER-4767131 SNASH@ENTERPRISE-KNOWLEDGE.COMJHILGER@ENTERPRISE-KNOWLEDGE.COM WWW.LINKEDIN.COM/IN/SARA-G-NASH Joseph Hilger Sara Nash @EKCONSULTING

×