Method for predicting citation counts

• US 20090157585A1
• Filed: 11/07/2008
• Published: 06/18/2009
• Est. Priority Date: 05/14/2004
• Status: Active Grant

• Current Assignee: Constantin Aliferis, Lawrence Fu
• Original Assignee: Constantin Aliferis, Lawrence Fu
• Inventors: Fu, Lawrence, Aliferis, Constantin
• Granted Patent: US 8,527,442 B2
• Time in Patent Office: Days
• US Class Current: 706/46
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 5/02 Knowledge representation; Symbolic representation

Abstract

A computerized process to predict citation counts of articles comprising the steps of receiving an article through an input, obtaining, through the input, a selected set of articles exclusive of the article, storing in a memory the set of articles and the article, extracting through a computer processor an article feature from each article in the stored set of articles, constructing models through the computer processor using a pattern recognition process and the article feature, selecting, through the processor, a best model, predicting by application of the best model to the article by the processor a future citation count of the article, outputting, the article comprising the future citation count and controlling through a publication controller unit, distribution of the article.

Claims

1. A computerized process to predict citation counts of articles comprising the steps of:

receiving an article through an input;

obtaining, through said input, a selected set of articles exclusive of said article;

storing in a memory said set of articles and said article;

extracting through a computer processor an article feature from each article in said stored set of articles;

constructing models through said computer processor using a pattern recognition process and said article feature;

selecting, through said processor, a best model;

predicting by application of said best model to said article by said processor a future citation count of said article;

outputting, said article comprising said future citation count; and

controlling through a publication controller unit, distribution of said article.

2. A computerized process according to claim 1, wherein said predicting by application of said best model step is done before said article receives any citation.

3. A computerized process according to claim 1, wherein said predicting by application of said best model step is done before said article is published.

4. A computerized process according to claim 1, wherein said article feature comprises at least one of an abstract, a title, a body, and a database identifier.

5. A computerized process according to claim 1 further comprising:

obtaining, through said input, an author feature for each article in said set of articles;

storing in said memory said author feature;

wherein said constructing, through said processor step uses said stored author feature.

6. A computerized process according to claim 5, wherein the author feature comprises at least one of a number of received citations of at least one author, a number of articles by at least one author, and a quality of at least one author'"'"'s institution.

7. A computerized process according to claim 1 further comprising:

obtaining, through said input for receiving, a bibliometric feature for each article in said set of articles;

storing in said memory said bibliometric feature;

wherein said constructing, through said processor step uses said stored bibliometric feature.

8. A computerized process according to claim 7, wherein said bibliometric feature includes at least one of a journal impact factor, a publication type, a number of authors, and a number of institutions.

9. A computer system for predicting a future citation count of an article and controlling publication based on said prediction comprising:

an input device for receiving information regarding an article;

a memory for storing said article, a set of articles, a bibliometric feature, and an author feature;

a processor configured to;

extract an article feature from each article in said stored set of articles, construct models using a pattern recognition process, said article feature, said stored author feature and said stored bibliometric feature, select a best model, predict, using said best model, said future citation count for said article;

an output device for outputting said article comprising said future citation count; and

a publication controller unit for controlling distribution of said article.

10. A computer system according to claim 9, wherein said bibliometric feature comprises at least one of a journal impact factor, a publication type, a number of authors, and a number of institutions.

11. A computer system according to claim 9, wherein said author feature comprises at least one of a number of received citations of at least one author, a number of articles by at least one author, and a quality of at least one author'"'"'s institution.

12. A computer system according to claim 9, wherein said article feature comprises at least one of an abstract, a title, a body, and a database identifier.

13. A computer readable medium containing a program for operating a computer for predicting a future citation count of an article, the program causing the computer to perform acts comprising:

obtaining, through an input, an article;

obtaining, through said input, an author feature for a selected set of articles exclusive of said article;

storing in a memory said author feature and said article;

constructing, through a processor, models using a pattern recognition process and said stored author feature;

selecting, through said processor, a best model;

predicting by application of said best model to said article by said processor a future citation count of said article;

outputting, through an output said article comprising said future citation count; and

controlling through a publication controller unit, distribution of said article.

14. A computer readable medium according to claim 13, wherein the author feature comprises at least one of a number of received citations of at least one author, a number of articles by at least one author, and a quality of at least one author'"'"'s institution.

15. A computer readable medium according to claim 13, further comprising:

extracting through said processor an article feature from each article in said set of articles;

wherein said constructing, through said processor uses said article feature.

16. A computer readable medium according to claim 15, wherein the wherein said article feature comprises at least one of an abstract, a title, a body, and a database identifier.

17. A computer readable medium according to claim 13, further comprising:

obtaining, through said input, a bibliometric feature for each article in said set of articles;

storing in said memory said bibliometric feature;

wherein said constructing, through said processor step uses said stored bibliometric feature.

18. A computer readable medium according to claim 17, wherein said bibliometric feature includes at least one of a journal impact factor, a publication type, a number of authors, and a number of institutions.