Method and system for determining classification of text

A text classification system for determining classification of text, comprises:

• a processor;
• and
• a memory communicatively coupled to the processor, wherein the memory stores processor-executable instructions, which, on execution, cause the processor to:
• receive a text from a plurality of texts;
• perform a training of the text classification system using a plurality of similar pair of texts and a plurality of dissimilar pair of texts from the plurality of texts, to obtain first trained model parameters of a first Long Short Term Memory (LSTM) unit and second trained model parameters of a second LSTM unit;
• generate a first vector representation of the text using the first trained model parameters of the first LSTM unit and a second vector representation of the text using the second trained model parameters of the second LSTM unit;
• combine the first vector representation and the second vector representation using a combiner operator to obtain a combined vector representation, wherein the combiner operator is selected from a plurality of combiner operators based on the training using an accuracy value of a classifier of the text classification system, wherein the accuracy value of the classifier is determined by comparing a classification performed by the classifier with a predefined classification stored in a data source associated with the text classification system;
• and
• provide the combined vector representation to the classifier for determining a classification of the text.

A non-transitory computer readable medium including instructions stored thereon that well processed by at least one processor cause a device to perform operations comprising:

• receiving, by a text classification system, a text from a plurality of texts;
• training the text classification system using a plurality of similar pair of texts and a plurality of dissimilar pair of texts from the plurality of texts, to obtain first trained model parameters of a first Long Short Term Memory (LSTM) unit and second trained model parameters of a second LSTM unit;
• generating, by the text classification system, a first vector representation of the text using the first trained model parameters of the first LSTM unit and a second vector representation of the text using the second trained model parameters of the second LSTM unit;
• combining, by the text classification systems, the first vector representation and the second vector representation using a combiner operator to obtain a combined vector representation, wherein the combiner operator is selected from a plurality of combiner operators based on the training using an accuracy value of a classifier of the text classification system, wherein the accuracy value of the classifier is determined by comparing a classification performed by the classifier with a predefined classification stored in a data source;
• and
• providing, by the text classification system, the combined vector representation to the classifier for determining a classification of the text.