IBM Watson and natural language processing
1 like652 views
IBM Watson Natural Language Processing allows developers to extract insights from unstructured text to build cognitive apps. It uses techniques like entity extraction, sentiment analysis, keyword extraction and more to analyze text and understand the underlying concepts, entities, and sentiment. Some example uses are to extract people, places and entities from news articles or to gather business intelligence from sources like social media, websites or documents to power applications in areas like customer care, business intelligence, and content recommendation.
IBM Watson and natural language processing
- 1. IBM Watson Natural LanguageProcessing Conferenza Intelligenza Artificiale, Politecnico di Milano e AICT – Milano, 5 ottobre 2017 Ing. Roberto Villa Manager of Human Centric practice and Research ecosystem, IBM
- 2. The biggest taxi company owns no cars. The largest accommodation company owns no real estate. The biggest media company owns no content. The largest retailer carries no inventory. VIDEO: https://www.youtube.com/watch?v=P18EdAKuC1U
- 3. Three capabilities differentiatecognitive systems from traditionalprogrammedcomputingsystems… Reasoning They reason. They understand underlying ideas and concepts. They form hypothesis. They infer and extract concepts. Learning They never stop learning getting more valuable with time. Advancing with each new piece of information, interaction, and outcome. They develop “expertise”.Understanding Cognitive systems understand like humans do. …. allowingthem to interactwith humans.
- 4. 4 Examples include: Analyst reports tweets Wire tap transcripts Battlefield docs E-mails Texts Forensic reports Newspapers Blogs Wiki Court rulings International crime database Stolen vehicle data Missing persons data Data, information, and expertise create the foundation. Cognitive systems rely on collections of data and information:
- 5. Retrieve and Rank 5 Entity Extraction Sentiment Analysis Emotion Analysis (Beta) Keyword Extraction Concept Tagging Taxonomy Classification Author Extraction Language Detection Text Extraction Microformats Parsing Feed Detection Linked Data Support Concept Expansion Concept Insights Dialog Document Conversion Language Translation Natural Language Classifier Personality insights Relationship Extraction Retrieve and Rank Tone Analyzer Emotive Speech to Text Text to Speech Face Detection Image Link Extraction Image Tagging Text Detection Visual Insights Visual Recognition AlchemyData News Tradeoff Analytics 50 underlying technologies …and then leverage Watson APIs to apply cognitivecapabilities Natural Language Classifier Tone Analyzer
- 6. Watson Narrative 6 Watson Natural Language Understanding
- 7. Watson Narrative THE SOLUTION: THE IBM WATSON NATURAL LANGUAGE SERVICE UNDERSTANDING ENABLES DEVELOPERS TO EXTRACT INSIGHTS FROM UNSTRUCTURED TEXT TO POWER A NEW GENERATION OF COGNITIVE APPS. IBM Watson Natural Language Understanding7 PROBLEM: INABILITY TO MINE UNSTRUCTURED DATA AND LACK OF SKILLS IN MACHINE LEARNING. Unstructured data can come in many forms: > free format text fields > picture > video > audio Text analytics deals with free format text fields, from: > email content > a web page > a word document > a pdf file
- 8. Watson Narrative IBM Watson Natural Language Understanding IBM WATSON NATURAL LANGUAGE UNDERSTANDING Natural language processing for advanced text analysis A sophisticated suite of natural language processing capabilities to analyze text and extract meta-data from content such as concepts, entities, keywords, categories, sentiment, emotion, relations, semantic roles, with options for customization to specific industries and domains. Example Use Cases Extract people, places, companies and other entities mentioned in a news article or text Demo: https://natural-language-understanding-demo.mybluemix.net/ Documentation: https://www.ibm.com/watson/developercloud/doc/natural-language-understanding/
- 9. Watson Narrative HOW IT WORKS IBM Watson Natural Language Understanding makes value-driven decisions easy by giving you the full story behind all of your data.
- 10. Watson Narrative FEATURES Keywords: Determine the most important keywords in your content. Concepts: Identify general concepts in your content. Categories: Categorize your content into a hierarchical 5-level taxonomy. Entities: Detect important people, places, geopolitical entities and other types of entities. Sentiment: Determine whether your content conveys positive or negative sentiment. 10 Emotion: Detect emotions such as anger, disgust, fear, joy or sadness that are conveyed by your content. Relations: Identify relationships between entities in your content. Semantic roles: Identify the subjects of actions, and the objects that they act upon. Metadata: Get author information, publication date and the title of your text or HTML content. Custom models: Use IBM Watson Knowledge Studio to collaborate on the creation of custom annotation models.
- 18. Watson Narrative HOW CAN I APPLY WATSON NATURAL LANGUAGE UNDERSTANDING? Business intelligence: How can I gather business intelligence from unstructured text to create dashboards and reports? Social media monitoring: How do I extract insights from monitoring social media? Content recommendation: How can I recommend content that readers might like? Brand management: How can I know what consumers are saying about my brand? Advertising optimization: Where should I place my ads so that I get to the right audience? IBM Watson Natural Language Understanding18 Customer Care & Brand Management Problem: Inability to quantify/extract meaning from social media and customer feedback Value: Improve customer satisfaction and loyalty, Improve decision making Business & Competitive Intelligence Problem: Inability to use unstructured internal/external data for decision making Value: Reduction in staff/improved productivity Clustering (Matching, Ad- tech, Recommendation) Problem: Inability to match offers, people, advertising and other content to interests and intent Value: Improved conversions and time on site
- 19. Extract insight from Social Media with Watson Natural Language Understanding VIDEO: https://www.youtube.com/watch?v=P18EdAKuC1U
- 20. Extract insight from Social Media with Watson Natural Language Understanding Based on CDC gov. data ( Center for Disease Control)
- 21. Extract insight from Social Media with Watson Natural Language Understanding The institute’s NLUsystemat Osnabrück Universityexamines approximately 500millionEnglishtweetsworldwide everyday