Knowledge Engineering
The application of knowledge engineering has exceeded specialized expert and decision support systems (DSS), with advancements in knowledge representation and reasoning (KRR) leading to general-purpose knowledge bases and powerful inference engines – all so computers can answer unforeseen questions with human knowledge.
Drug discovery white paper
Read up on accelerating drug discovery with publicly available datasets such as SemMed and the Human Protein Atlas.
Drug discovery webinar
Learn how to model bioactive and biopharmaceutical data, and use automated inference to discover new drug uses.
Cosmos presentation
Watch a presentation from Vaticle's worldwide conference on building a clinical decision support system.
Requirements
Knowledge engineers need to integrate data which spans multiple domains, including from different lines of business, transform it into information defined by categories, names and relations, and create rules that reason over knowledge via logic.
Drug Discovery
Find new drugs faster by examining the relationships between diseases, pathways, genes, proteins and drugs to accelerate target identification, drug repurposing and disease gene prioritization.
Risk Management
Identify, assess, control and mitigate business risks (e.g., financial or reputational) by analyzing any and all connections between a business and potential threats, direct or indirect.
Supply Chain Management (SCM)
Improve the speed and efficiency of supply chains, and alleviate or prevent disruptions, by optimizing the connections between stores, warehouses, factories, suppliers and more.
Data Fabric
Create a unified layer that connects all of a business's data, often sourced from disparate databases, so employees can discover and access knowledge which encompasses the entire business.
Challenges
The biggest hurdle to effective knowledge engineering is connecting otherwise independent datasets in a way that makes it easy for others to find previously unknown information derived from new connections – whether direct, indirect or inferred.
Ongoing learning
The problem with constant knowledge acquisition is the disruption caused by having to continuously update the database schema in order to add knowledge requiring new types of data.
Endless connections
It's hard to grow a complete domain of knowledge because multilayered, multidimensional and highly interconnected data is exceedingly difficult to model and query.
Complex labyrinths
Once domain knowledge has been mapped, navigating the data and finding relevant paths among an unknown number of complex permutations is virtually impossible to manually.
Diverse datasets
It's common for the ideal knowledge base to include datasets from multiple sources, public and/or private, but traditional databases are designed to link disparate data.
TypeDB Solution
TypeDB provides cybersecurity software and teams with a database capable of storing every detail and connection that exists within their domain, both digital and physical, and applying logic and reason to help extrapolate critical information from it.
Extensible type systemAdd new types of infrastructure, threats, attacks and users on demand, upon discovery and without having to modify existing queries. | |
Inheritance and polymorphismQuery on abstract types (e.g., all injection attacks) rather than having to know and specify all possible types (e.g., SQL injection) with joins. | |
Relations as a core typeIdentify the connections between attackers and assets, from assets reachable by a compromised server to the next steps in a complex attack. | |
Built-in reasoningLet the reasoning engine traverse relations to find relevant information (e.g., assets a user has indirect access to) without specifying them. |
Further Learning
Download sample projects to get up and running in minutes, and check out the latest blogs from our research engineers.
Drug discovery white paper
Read up on accelerating drug discovery with publicly available datasets such as SemMed and the Human Protein Atlas.
Drug discovery webinar
Learn how to model bioactive and biopharmaceutical data, and use automated inference to discover new drug uses.
Drug discovery project
Download the TypeDB Bio project to create a core drug discovery schema, load open-source data, and infer new drug targets.