System and method for generating summary of research document
Disclosed is a system for generating summary of at least one research document. The system comprising computing device associated with an entity, data repository comprising ontological database and synonym database and server arrangement communicably coupled via one or more data communication networks with the computing device and the data repository. The server arrangement is configured to acquire information included in at least one research document, analyze information using ontological database and synonym database to identify set of keywords corresponding to at least one research document, assign first score to each of the keywords based on document-centric property, assign second score to one or more relationships between the keywords based on relationship-centric property and generate summary for at least one research document. Summary comprises: first portion generated based upon informatory data, second portion generated based on keywords and third portion generated using machine learning algorithm based on first score of keywords.
1. A system for generating a summary of at least one research document, the system comprising:
a computing device associated with an entity, wherein the computing device, comprises a computer readable program code, configured to:
upload the at least one research document,
acquire informatory data related to the at least one research document, and
preprocess the at least one research document to extract information included;
a data repository comprising an ontological database and a synonym database; and
a server arrangement communicably coupled via one or more data communication networks with the computing device and the data repository, the server arrangement configured to:
acquire, from the computing device, the information included in the at least one research document,
analyze, the information using the ontological database and the synonym database to identify a set of keywords corresponding to the at least one research document,
assign a first score to each of the keywords based on a document-centric property, the informatory data and a popularity index of each of the keyword, wherein the popularity index of each of the keyword is a metric for quantifying number of times the keyword is included in a web-activity, and wherein the document-centric property of a keyword includes at least one of: a location of the keyword in the at least one research document, an occurrence-frequency of the keyword in the at least one research document,
assign a second score to one or more relationships between the keywords based on a relationship-centric property, wherein assigning the second score to one or more relationships between the keywords, based on the relationship-centric property, includes:
identifying one or more relationships between the keywords,
identifying semantics of the one or more relationships in the at least one research document, and
analyzing world knowledge to determine a cognizance-index of the semantics of each of the one or more relationships, wherein the cognizance-index denotes an awareness of the one or more relationships, and
generate the summary for the at least one research document, wherein the summary comprises:
a first portion generated based upon the informatory data,
a second portion generated based on the keywords in the one or more relationships having the second score below a predefined threshold, and
a third portion generated, using a machine learning algorithm, based on the first score of the keywords.
2. The system according to the claim 1 , wherein preprocessing includes extracting entire content pertaining to the at least one research document.
3. The system according to the claim 1 , wherein preprocessing includes extracting selective content pertaining to the at least one research document.
4. The system according to the claim 1 , wherein the informatory data includes:
metadata related to the at least one research document,
hypotheses of the at least one research document, and
statistical significance of the hypotheses.
5. The system according to the claim 1 , wherein the machine learning algorithm is implemented as a natural language generator.
6. A method for generating a summary of at least one research document, wherein the method is implemented using a system comprising:
a computing device associated with an entity, wherein the computing device, comprises a computer readable program code, configured to:
upload the at least one research document,
acquire informatory data related to the at least one research document, and
preprocess the at least one research document to extract information;
a data repository comprising an ontological database and a synonym database; and
a server arrangement communicably coupled via one or more data communication networks with the computing device and the data repository, wherein the method comprises:
acquiring, from the computing device, the information included in the at least one research document,
analyzing, the information using the ontological database and the synonym database to identify a set of keywords corresponding to the at least one research document,
assigning a first score to each of the keywords based on a document-centric property, the informatory data and a popularity index of each of the keyword, wherein the popularity index of each of the keyword is a metric for quantifying number of times the keyword is included in a web-activity, and wherein the document-centric property of a keyword includes at least one of: a location of the keyword in the at least one research document, an occurrence-frequency of the keyword in the at least one research document,
assigning a second score to one or more relationships between the keywords based on a relationship-centric property, wherein assigning the second score to one or more relationships between the keywords, based on the relationship-centric property, includes:
identifying one or more relationships between the keywords,
identifying semantics of the one or more relationships in the at least one research document, and
analyzing world knowledge to determine a cognizance-index of the semantics of each of the one or more relationship, wherein the cognizance-index denotes an awareness of the one or more relationships, and
generating the summary for the at least one research document, wherein the summary comprises:
a first portion generated based upon the informatory data,
a second portion generated based on the keywords in the one or more relationships having the second score below a predefined threshold, and
a third portion generated, using a machine learning algorithm, based on the first score of the keywords.
7. The method according to the claim 6 , wherein preprocessing includes extracting entire content pertaining to the at least one research document.
8. The method according to the claim 6 , wherein preprocessing includes extracting selective content pertaining to the at least one research document.
9. The method according to the claim 6 , wherein the informatory data includes:
metadata related to the at least one research document,
hypotheses of the at least one research document, and
statistical significance of the hypotheses.