IBMtoolmapping&roles.txt

Data Science Methodology - IBM Tool mapping & roles:

Analytic Approach: 
- ML: Watson Machine Learning
- Statistical: SPSS Statistics/Modeler found in Watson Studio

Data Requirements:

Data Tools chosen:

Data Collection:

Data Understanding:

Data Preprocessing(Cleansing, Integration, Transformation, Reduction):
- Watson Studio (Data Refinery for cleansing, transformation) self service

Data Modeling:
- Watson Studio (can use Anaconda(Python), Spark(Big Data), GPU(Deep Learning) for testing and deploying AI models) self service        includes SPSS Statistics/Modeler (descriptive) &  Decision Optimization(prescriptive)
- NLU API - create a custom model using NLP for text analytics for some APIs to get specific results that are tailored to your domain
- WKS API - 
- WDS API - (use WKS to annotate & build/deploy custom model specific for your domain)
- Watson Explorer


Data Evaluation:

Data Model Deployment/Updating via Feedback:
- Watson Machine Learning
- Watson Studio (can use Anaconda(Python), Spark(Big Data), GPU(Deep Learning) for testing/trained and deploying AI models) 
  and SPSS Statitics/Modeler for statistical model deployment self service


Post Deployment
- AI openscale integrates with WML (analyze deployed models at scale, transparency into decisions/predictions, Continuously deploy customized, AI generated neural networks without code)

Infuse AI:
Watson Assistant-Customer service
Watson Discovery-data discovery for knowledge workers
Natural Language Understanding
Visual Recognition
Speech-To-Text
Text-To-Speech