CognitiveComputing_CirrusShakeri_final
3 likes890 views
Cognitive computing systems learn about a domain by processing large amounts of data and text to automatically encode knowledge. Technologies like natural language processing, machine learning, and reasoning enable interactions through conversation and applications that discover relationships to interact like humans. Building cognitive applications involves mapping data landscapes, selecting an application, listing required functions, building a knowledge graph, implementing algorithms, automating data ingestion, and iterating with subsets of data.
CognitiveComputing_CirrusShakeri_final
- 2. What is Cogni,ve Compu,ng? • Programming Model – A Cogni,ve Compu,ng system learns about a domain (e.g., medical diagnos,cs) by processing large amounts of data and unstructured text from medical literature and medical databases. • Technology Components – Cogni,ve Compu,ng is comprised of a set of technologies that automate the encoding of large amounts of data into a computer-‐understandable knowledge base. Examples of technologies in Cogni,ve Compu,ng include natural language understanding, machine learning, and probabilis,c and seman,c reasoning. • User Interac,on – Cogni,ve Compu,ng systems engage in conversa,ons with their human users in natural language by asking and answering ques,ons. • Applica,ons – Cogni,ve Compu,ng enables machines to reason about the objects and events in a specific domain of discourse, and to discover rela,onships among things in that domain. This enables applica,ons that can interact conversa,onally with humans as if they were human.
- 3. 3 THE PATH TO COGNITIVE COMPUTING PASSES THROUGH (BIG) DATA
- 4. Big Data Data Ingestion Knowledge Graph Machine Reasoning Cognitive Function Web Content (web sites, blogs, …) Predict (demand, inventory, …) Learning from Usage Patterns Semantic Inferencing Big Data ! Intelligent Applications (A Lifecycle View) Social Networks (twitter, facebook, …) Enterprise Apps (ERP, CRM, …) Internet of Things (sensor data, device data, …) Textual Content (documents, reports, …) Online Activities (search, shopping, …) Knowledge-bases (taxonomies, ontologies, …) Data Preparation • Data integration • Data enrichment • Data imputation • Data versioning • Data provenance • … Natural Language Processing • Entity extraction • Entity resolution • Relationship extraction • Taxonomy generation • Knowledge based population (slot filling) • … Context Engine Sensemaking Engine Semantic Search Machine Learning (classification, clustering, anomaly detection, …) Design (product, process, …) Analyze (performance, problem, …) Detect (incident, anomaly, opportunity, …) Find (people, content, …) Discover (insight, pattern, …) Compare (products, companies,, …) Processes (process logs, server logs, …) Automated Update Cycle Rule Engine Process Automation Engine Semantic Query Engine Inference Engine Network of: people, places, organizations, processes, rules, policies, events, documents, devices, … Recommendation Engine Cirrus Shakeri, Inventurist LLC – All Rights Reserved …… …
- 5. How to Build Cogni,ve Compu,ng Applica,ons (in the enterprise) 1 MaptheDataLandscape 2 Ideationand Selectionofthe Application 3 ListCognitive FunctionsRequired Do you have the right datasets? 4a BuildtheKnowledge Graph 4b ImplementAlgorithms 5 AutomateData Ingestion Iterate with subset of data
- 6. 6 COGNITIVE COMPUTING TO DRIVE THE STARTUP ECONOMY An Example
- 7. From Data to Intelligence Knowledge Graph Data Knowledge Graph Results Twitter CrunchBase Company web site Metatag Nearest neighbor Shortest path Spreading activation Competitive landscape Sub-graph distance Investors active in this domain ... Potential customers ... Large corporations ... 7
- 8. Big Data From Data to Intelligence to Ac,on Intelligence Action 8
- 9. Thank You 9 Contact Cirrus Shakeri, Inventurist • cirrus.shakeri@inventurist.com • twiWer.com/cirrus_shakeri • www.linkedin.com/in/cshakeri/