B.Tech Computer Science with Specialization in Knowledge Engineering

Motivation

Computing power is growing exponentially and becoming ubiquitous and more affordable as a utility. Recent advances in storage and data management technologies are enabling a revolution in the amount of data that can be collected, stored and processed in real-time. These two trends have together spurred new growth in Big Data algorithms and their application for information processing. The consequence is revival of artificial intelligence where systems mimic human intelligence to perform tasks considered impractical until recently – such as self-driving cars, autonomous robots and drones.

The next evolutionary step is machine-processing of knowledge and intelligent human-like interfaces – all in natural languages. Though several enabling technologies have been around for a while, they are now becoming affordable at scale to create useful applications.

What is Knowledge Engineering?

Knowledge engineering refers to modeling, storing, processing, mining, correlation and inference of knowledge typically embedded in natural languages in various media. This field has been in existence for decades in the form of expert systems, though at small scale for niche applications. However, it is re-emerging as a growth area in computing after Big Data analytics. It is especially relevant to India now due to the government’s thrust on language-enabled services and exploiting indigenous knowledge for sustainable growth.

Applications of Knowledge Engineering

India has one of the richest repositories of human knowledge preserved over millennia in languages such as Sanskrit that are naturally more amenable to machine-processing. The bulk of them are in oral traditions as well as in 4+ million written manuscripts in dozens of Indic languages and scripts. They are yet to be deciphered, and span a wide variety of subjects including ecology, biology, psychology, agriculture, metallurgy, architecture and astronomy. Mining this knowledge base requires applying the latest computing algorithms and technologies in innovative ways. Moreover, India’s traditional sciences called pada-vākya-pramāṇaśāstra-s extensively deal with language interpretation and precise and concise knowledge representation. Many śāstra texts adhere to their rules, making them more amenable to machine-processing than modern language texts.

The MIT ADT B.Tech Computer Science Program

MIT ADT University, Pune offers a standard B.Tech Computer Science program, but with a novel choice to learn concepts and techniques of knowledge engineering in the context of Indic knowledge systems Studying knowledge engineering in the Indic context is beneficial to students in two ways.

  1. First-hand experience in applying computer science principles and data analytics to a real-world problem, namely, decoding India’s vast knowledge base.
  2. Rigorous exposure to India’s native contributions in linguistics and knowledge sciences and their application to modern knowledge engineering.

Syllabus for Knowledge Engineering Specialization

The following courses will be administered by MIT ADT University College of Engineering, Pune in addition to the standard B.Tech Computer Science curriculum as stipulated by AICTE. These courses will be taught by the faculty of MIT ADT’s School of Vedic Sciences during the semesters mentioned.

Semester

Course Name

Topics Covered

I Year, II Semester

The Science of Effective Reasoning (common to all MIT ADT programs)

Indian & western systems of logic, debate (Nyaaya) and knowledge representation

II Year, II Semester

Saṁskṛta Computational Linguistics

Introduction to Natural Language Processing, syntax & semantic analysis of Indic languages esp. Saṁskṛtam, computational structure of Pāṇini’s Aṣṭādhyāyī

III Year, II Semester

Elements of Knowledge Processing

Semantic networks, discourse analysis, text mining, logic programming

IV Year, II Semester

Final Project

Indic knowledge processing

Skills Acquired

In addition to the skills acquired through a standard B.Tech Computer Science program, the knowledge engineering specialization imparts the following additional skills through hands-on projects:

  • Horizontal Technologies: Graph algorithms/databases, inference, neural networks, cloud-based scale-out architectures
  • Vertical Technologies: Image/speech processing, text-mining and/or computational linguistics
  • Contemporary Skill Sets: Python programming, RESTful services, WebUI development (MVC model), NoSQL databases, ...

Career Options

  • Enhanced employability via hands-on experience with industry-valued technologies
  • Competitive edge in an emerging area of language and knowledge processing
  • Jump start to pursue higher studies in cutting edge computer science