Automated extraction and classification of project initiation related information from request for proposal response documents

A system, comprising:

• a memory storing instructions;
• one or more communication interfaces;
• a question and answer extractor;
• one or more classifiers;
• a project initiation kit generator;
• a Natural Language Processing (NLP) engine;
• and
• one or more hardware processors coupled to the memory via the one or more communication interfaces, wherein the one or more hardware processors are configured by the instructions to:
• receive a Request for Proposal (RFP) response document and a configurable extraction pattern as inputs, wherein the configurable extraction pattern specifies a heading and style related information of the RFP response document;
• extract a plurality of questions and a plurality of corresponding answers by analyzing the Request for Proposal (RFP) response document based on the configurable extraction pattern;
• classify each of the plurality of questions into at least one project initiation class of a set of predefined project initiation classes based on a comparison of a configurable threshold value with a probability score computed using a pre-trained machine learning model;
• process each of the plurality of corresponding answers to extract a plurality of project initiation related information using the one or more classifiers, wherein the plurality of project initiation related information is extracted by processing text information and images present in each of the plurality of corresponding answers;
• generate a plurality of class specific documents with the plurality of project initiation related information;
• and
• compose one or more stakeholder specific documents involved in project initiation process based on the plurality of class specific documents, wherein the one or more stakeholder specific documents are further shared with corresponding one or more stakeholders for project related resource planning.

One or more non-transitory machine-readable information storage mediums comprising one or more instructions which when executed by one or more hardware processors cause:

• receiving, a Request for Proposal (RFP) response document and a configurable extraction pattern as inputs, wherein the configurable extraction pattern specifies a heading and style related information of the RFP response document;
• extracting, via the one or more hardware processors, a plurality of questions and a plurality of corresponding answers by analyzing the Request for Proposal (RFP) response document based on the configurable extraction pattern;
• classifying, via the one or more hardware processors, each of the plurality of questions into at least one project initiation class of a set of predefined project initiation classes based on a comparison of a configurable threshold value with a probability score computed using a pre-trained machine learning model;
• processing, via the one or more hardware processors, each of the plurality of corresponding answers to extract a plurality of project initiation related information using one or more classifiers, wherein the plurality of project initiation related information is extracted by processing text information and images present in each of the plurality of corresponding answers;
• generating, via the one or more hardware processors, a plurality of class specific documents with the plurality of project initiation related information;
• and
• composing, via the one or more hardware processors, one or more stakeholder specific documents involved in project initiation process based on the plurality of class specific documents, wherein the one or more stakeholder specific documents are further shared with corresponding one or more stakeholders for project related resource planning.