Activity 11: Build a Text Summarization Model

We will use the attention mechanism model architecture we built for neural machine translation to build a text summarization model. The goal of text summarization is to write a summary of a given large text corpus. You can imagine using text summarizers for the summarization of books or the generation of headlines for news articles.

As an example, use the given input text:

"Celebrating its 25th year, Mercedes-Benz India is set to redefine India's luxury space in the automotive segment by launching the new V-Class. The V-Class is powered by a 2.1-litre BS VI diesel engine that generates 120kW power, 380Nm torque, and can go from 0-100km/h in 10.9 seconds. It features LED headlamps, a multi-functional steering wheel, and 17-inch alloy wheels."

A good text summarization model should be able to produce a meaningful summary, such as:

"Mercedes-Benz India launches the new V-Class"

From an architectural viewpoint, a text summarization...