Custom Classifier 2.0

We have built this classifier for text classification which relies on Zero-Shot learning technique called as Custom Classifier. Our base model is trained on a large news corpus of 10 million news articles to discover relationships between sentences and their categories. The resulting model can then generalize on new, unseen categories as well not available in training data.

Custom Classifier 2.0 is now called SmartReader. SmartReader is based on deep learning where AI trains your data to automatically give you topics and sub-topics from your data. This data is further analyzed and gives confidence score for each topic. SmartReader has some salient features such as smart topic identification, automatic categorization, and in-depth analysis of data. SmartReader can quickly analyze large sets of open-ended verbatim comments, surveys, feedback and find actionable insights to boost your business.

Create Your Free Model

try our custom classifier 2.0 demo

enter a text

In the demo below, provide an input text and some categories (labels) in which you want to classify the text.

add your categories

Why our Custom Classifier 2.0 API ?


Highly accurate classification of unstructured textual data.

Real Time

State of the art technology to provide accurate results real-time.


Can be trained on custom dataset to obtain similar accuracy and performance.

How Our Custom Classifier 2.0 API Works?

Zero Shot Learning is a way to be able to infer dataset's members without training on it. It is mostly achieved by some form of transfer learning, by which knowledge learned from one dataset can be applied on a different one. While people have proposed multiple zero shot learning approaches for vision tasks where knowledge from imagenet dataset can be used on new ones, we haven't yet seen any example of zero shot learning for text classification.

In our latest research work, we have proposed a method to do zero shot learning on text, where an algorithm trained to learn relationships between sentences and their categories on a large noisy dataset can be made to generalize to new categories or even new datasets. We call the paradigm "Train Once , Test Anywhere". We also propose multiple neural network algorithms that can take advantage of this training methodology and get good results on different datasets. The best method uses an LSTM model for the task of learning relationships. The idea is if one can model the concept of "belongingness" between sentences and classes, the knowledge is useful for unseen classes or even unseen datasets.

use cases

Analyze Verbatim Comments

Verbatim comments may pose a challenge at first but analyzing them can be made easy and simple with the help of Deep Learning. We have launched custom classifier 2.0 which is now SmartReader to allow anyone to analyze verbatim comments quickly and accurately without writing a single line of code. The algorithm deciphers the output of the data set by mining key phrases that are tracked to particular labels. Read more...

Customer Survey Analysis

There is a need to distinguish between positive and negative feedback while performing survey analysis. Negative feedback is a goldmine of potential product enhancements, and positive feedback tells you where you are adding the most value to your customers. Using the add-on makes it super easy to use AI and find out the underlying sentiment behind thousands of text pieces in one go. Read more...


(30-day subscription)

Get Started

Sign Up for a free account

Interested in integrating with our APIs? Sign up now.

Sign up
Custom Solutions

Want to train your own custom model? Contact Sales to get started

Create a Model