Cognitive Computing

In the name of Allah, most gracious and most merciful,

You can see the full Mind Map in more details by clicking here

1. Definition

1.1 Definition 1

IBM defines cognitive computing as an advanced system that learns at scale, reason with purpose, and interacts with humans in a natural form. They learn and reason from their interactions with us and from their experiences with their environment instead of being explicitly programmed.

It is not just about implementing the best algorithm for solving a problem, Cognitive Computing tries to mimic human intelligence by analyzing a series of factors. It learns by studying patterns and imitates the human thinking process.

Instead of completing the task independent of humans by implementing a pre-defined algorithm, it serves as an assistant to humans in decision-making by suggesting humans to take relevant action. Therefore it gives humans the power of computers fast and accurate data analysis without having to worry about the machine learning system wrong decisions.

To be more clear, cognitive computing is computing that is more focused on reasoning and understanding at a higher level. It could deal with symbolic and conceptual information instead of just pure data or sensor streams.

Cognitive computing goes well beyond artificial intelligence and human-computer interaction as we know it–it explores the concepts of perception, memory, attention, language, intelligence and consciousness. Typically, in AI, one creates an algorithm to solve a particular problem. Cognitive computing seeks a universal algorithm for the brain. This algorithm would be able to solve a vast array of problems.

Dharmendra Modha, Manager of Cognitive Computing at IBM Research

1.1 Definition 2

Cognitive Computing refers to technology platforms that are based on artificial intelligence and signal processing.

Possible features of Cognitive Systems:

  1. Adaptive
    • Learning as information changes, and as goals and requirements evolve.
    • Resolving ambiguity and tolerate unpredictability.
    • May feed on dynamic data in real time, or near real time.
  2. Interactive
    • Easily interaction with users so that users could define their needs easily.
    • Interaction with other processors, devices, and cloud services.
  3. Iterative and Stateful
    • May help in defining problems by asking questions or finding additional input incase of ambiguous or incomplete problem statement
    • May remember process previous interactions and return suitable information for a certain application at a that point in time.
  4. Contextual
    • May understand meaning, syntax, time, location, suitable domain, regulations, process, task, and goal.

2. Topics

2.1 Data Mining

Extracting and discovering anomalies, patterns, and correlations in large data sets using machine learning, statistics, and database systems to predict outcomes.

2.2 Pattern Recognition

Automated recognition of patterns in data.

2.3 Natural Language Processing “NLP”

Natural Language Processing came as a combination of computer science (including Artificial Intelligence), and linguistics with the aim of making computers understand the human language, and to analyze huge amounts of natural language data for building useful applications.

For more information, you could refer to this NLP post.

2.4 Signal Processing

It is a subfield of electrical engineering that focuses on analyzing, modifying, and synthesizing signals like sounds, and images.

What you Expect to Learn (Cognitive Computing Recipes Textbook (Artificial Intelligence Solutions Using Microsoft Cognitive Services and TensorFlow) Table of Contents)

  1. Democratization of AI Using Cognitive Services
    1. Democratization of Artificial Intelligence
      1. Machine-Learning Libraries
      2. Current State of Machine-Learning & Deep-Learning Platforms
    2. Building a Business Case for Artificial Intelligence
      1. Natural-Language Understanding & Generation
      2. Speech Recognition
      3. Cognitive Digital Assistants
      4. Unstructured Text Analytics
      5. Decision Management
      6. Robotic Process Automation
    3. The Five Tribes of Machine Learning
    4. Microsoft Cognitive Services—A Whirlwind Tour
      • Speech
    5. Language
    6. Knowledge
    7. Search
    8. Ethics of Artificial Intelligence
  2. Building Conversational Interfaces
    1. Components of Conversational UI
    2. Getting Started with Bot Framework
    3. Bot Framework SDK Samples
    4. Recipe 2-1. Building YodaBot
    5. Recipe 2-2. Creating Bots Using Azure Bot Service
    6. Recipe 2-3. Building a Question and Answer Bot
    7. Recipe 2-4. Data Center Health Monitor Bot
    8. Setting Up Azure Deployment via Resource Manager Template
  3. Seeing Is Believing: Custom Vision
    1. Hot Dog, Not Hot Dog
    2. Building Custom Vision to Train Your Security System
    3. Caption Bot Using the Cognitive Services Computer Vision API
    4. Explore Your Fridge Using CustomVision.AI
    5. Now Explore Your Fridge Using Cognitive Toolkit
    6. Product and Part Identification Using Custom Vision
    7. Apparel Search with Custom Vision Models in CNTK
  4. Text Analytics: The Dark Data Frontier
    1. Overview of the Text Analytics Ecosystem
    2. Claim Classification
    3. Know Your Company’s Health
    4. Text Summarization
  5. Cognitive Robotics Process Automation: Automate This!
    1. Extract Intent from Audio
    2. Email Classification and Triage for Automated Support-Ticket Generation
    3. Anomaly Detection: A Case of Fraudulent Credit Card Transactions
    4. Finding Needles: Cross-Correlation in Time Series
    5. Understanding Traffic Patterns: Demand Forecasting for Energy
  6. Knowledge Management & Intelligent Search
    1. Explore the Azure Search Indexing Process
    2. Natural-Language Search with LUIS
    3. Implement Entity Search
    4. Get Paper Abstracts
    5. Identify Linked Entities in Text Analytics
    6. Apply Cognitive Search
  7. AIOps: Predictive Analytics & Machine Learning in Operations
    1. Building Knowledge Graph Using Grakn
    2. Detect Anomalies Using Cognitive Services Labs Project Anomaly Finder
  8. AI Use Cases in the Industry
    1. Financial Services
      1. Mobile Fraud Detection
      2. Float Optimization
      3. Accident Propensity Prediction (Insurance)
    2. Healthcare
      1. Accurate Diagnosis and Patient Outcome Prediction
      2. Hospital Readmission Prediction and Prevention
    3. Automotive & Manufacturing
      • Predictive Maintenance
    4. Retail
      1. Personalized Storefront Experiences
      2. Fast Food Drive-thru Automation Problem
    5. Appendix A: Public Datasets & Deep Learning Model Repositories

3. Benefits

  • Accurate Data Analysis
  • Leaner & More Efficient Business Processes
  • Improved Customer Interaction

4. Applications

  • Smart Internet of Things “IoT”
  • AI-Enabled Cybersecurity
  • Next Generation Search
  • Content AI
  • Intent-Based Natural Language Processing
  • Cognitive Analytics in Healthcare

5. Skills required to work in Cognitive Computing

From my search, I didn’t find specific skills for that field. However, by viewing job descriptions I found that the skills required are mostly similar to Machine Learning skills with more focus on Natural language Processing skills.

So you could refer to the skills I have mentioned in that machine learning post, and that NLP post.


Thank you. I hope this post has been beneficial to you. I would appreciate any comments if anyone needed more clarifications or if anyone has seen something wrong in what I have written in order to modify it, and I would also appreciate any possible enhancements or suggestions. We are humans, and errors are expected from us, but we could also minimize those errors by learning from our mistakes and by seeking to improve what we do.

Allah bless our master Muhammad and his family.


Notify of

Most Voted
Newest Oldest
Inline Feedbacks
View all comments
3 years ago

All above told the truth.

Would love your thoughts, please comment.x