Method and system for topic-based classification of scientific papers to research proposal

A system comprising:

• a memory storing a plurality of instructions;
• one or more communication interfaces;
• 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 constructed dataset as an input data using at least one existing research paper, wherein the constructed dataset pertain to at least one of:
• (i) a research proposal title, or (ii) an abstract, or (iii) a set of proposal topics, or (iv) at least one reference paper pertinent to a research proposal, and (v) at least one citation text under at least one proposal topic from the set of proposal topics corresponding at least one relevant reference paper;
• augment the constructed dataset with at least one reference text span to obtain an extended dataset, wherein at least one reference text span pertain to at least one pseudo positive reference text span, and at least one pseudo negative reference text span, wherein at least one top-K chunk is identified from at least one reference paper which is relevant to at least one citation text based on a retriever model, is considered as at least one reference text span;
• train a topic-based retriever (TR) model on a subset of the extended dataset by at least one pseudo positive reference text span, and at least one pseudo negative reference text span;
• train a topic classifier (TC) model on a subset of the extended dataset using the research proposal title, at least one proposal topic, and at least one reference text span from at least one reference paper retrieved using at least one pseudo positive reference text span retrieved using the citation text based retriever model, which is labelled as positive, and at least one pseudo negative reference text span for at least one reference paper which is labelled as negative, to classy if at least one reference paper is aligned to the proposal topic or not;
• and
• classify at least one reference paper to be relevant to at least one proposal topic in context of at least one reference text span retrieved using the topic-based retriever (TR) model with the topic classifier (TC) model.

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 constructed dataset as an input data using at least one existing research paper, wherein the constructed dataset pertain to at least one of:
• (i) a research proposal title, or (ii) an abstract, or (iii) a set of proposal topics, or (iv) at least one reference paper pertinent to a research proposal, and (v) at least one citation text under at least one proposal topic from the set of proposal topics corresponding at least one relevant reference paper;
• augmenting the constructed dataset with at least one reference text span to obtain an extended dataset, wherein at least one reference text span pertain to at least one pseudo positive reference text span, and at least one pseudo negative reference text span, wherein at least one top-K chunk is identified from at least one reference paper which is relevant to at least one citation text based on a retriever model, is considered as at least one reference text span;
• training a topic-based retriever (TR) model on a subset of the extended dataset by at least one pseudo positive reference text span, and at least one pseudo negative reference text span;
• training a topic classifier (TC) model on a subset of the extended dataset using the research proposal title, at least one proposal topic, and at least one reference text span from at least one reference paper retrieved using at least one pseudo positive reference text span retrieved using the citation text based retriever model, which is labelled as positive, and at least one pseudo negative reference text span for at least one reference paper which is labelled as negative, to classy if at least one reference paper is aligned to the proposal topic or not;
• and
• classifying at least one reference paper to be relevant to at least one proposal topic in context of at least one reference text span retrieved using the topic-based retriever (TR) model with the topic classifier (TC) model.