Forecasting turning trends in knowledge networks

Ádám Szántó-Várnagy, I. Farkas

Research output: Contribution to journalArticle

Abstract

A large portion of our collective human knowledge is in electronic repositories. These repositories range from “hard fact” databases (e.g., scientific publications and patents) to “soft” knowledge such as news portals. The common denominator between them all is that they can be thought of in terms of topics/keywords. The interest in these topics is constantly changing over time. Their frequency occurrence diagrams can be used for effective prediction by the most straightforward simplification. In this paper, we use these diagrams to produce simple and human-readable rules that are able to predict the future trends of the most important keywords in 5 data sets of different types. A thorough analysis of the necessary input variables and parameters and their relation to the success rate is presented, as well.

Original languageEnglish
Pages (from-to)110-122
Number of pages13
JournalPhysica A: Statistical Mechanics and its Applications
Volume507
DOIs
Publication statusPublished - Oct 1 2018

Fingerprint

forecasting
Repository
Forecasting
Diagram
diagrams
Common denominator
trends
news
patents
Patents
simplification
Simplification
Electronics
occurrences
Predict
Necessary
Prediction
predictions
electronics
Range of data

Keywords

  • Prediction
  • Time dynamics
  • Topic evolution
  • Topic similarities

ASJC Scopus subject areas

  • Statistics and Probability
  • Condensed Matter Physics

Cite this

Forecasting turning trends in knowledge networks. / Szántó-Várnagy, Ádám; Farkas, I.

In: Physica A: Statistical Mechanics and its Applications, Vol. 507, 01.10.2018, p. 110-122.

Research output: Contribution to journalArticle

@article{ec9877cbed514b08a0bd49110e99f22a,
title = "Forecasting turning trends in knowledge networks",
abstract = "A large portion of our collective human knowledge is in electronic repositories. These repositories range from “hard fact” databases (e.g., scientific publications and patents) to “soft” knowledge such as news portals. The common denominator between them all is that they can be thought of in terms of topics/keywords. The interest in these topics is constantly changing over time. Their frequency occurrence diagrams can be used for effective prediction by the most straightforward simplification. In this paper, we use these diagrams to produce simple and human-readable rules that are able to predict the future trends of the most important keywords in 5 data sets of different types. A thorough analysis of the necessary input variables and parameters and their relation to the success rate is presented, as well.",
keywords = "Prediction, Time dynamics, Topic evolution, Topic similarities",
author = "{\'A}d{\'a}m Sz{\'a}nt{\'o}-V{\'a}rnagy and I. Farkas",
year = "2018",
month = "10",
day = "1",
doi = "10.1016/j.physa.2018.05.055",
language = "English",
volume = "507",
pages = "110--122",
journal = "Physica A: Statistical Mechanics and its Applications",
issn = "0378-4371",
publisher = "Elsevier",

}

TY - JOUR

T1 - Forecasting turning trends in knowledge networks

AU - Szántó-Várnagy, Ádám

AU - Farkas, I.

PY - 2018/10/1

Y1 - 2018/10/1

N2 - A large portion of our collective human knowledge is in electronic repositories. These repositories range from “hard fact” databases (e.g., scientific publications and patents) to “soft” knowledge such as news portals. The common denominator between them all is that they can be thought of in terms of topics/keywords. The interest in these topics is constantly changing over time. Their frequency occurrence diagrams can be used for effective prediction by the most straightforward simplification. In this paper, we use these diagrams to produce simple and human-readable rules that are able to predict the future trends of the most important keywords in 5 data sets of different types. A thorough analysis of the necessary input variables and parameters and their relation to the success rate is presented, as well.

AB - A large portion of our collective human knowledge is in electronic repositories. These repositories range from “hard fact” databases (e.g., scientific publications and patents) to “soft” knowledge such as news portals. The common denominator between them all is that they can be thought of in terms of topics/keywords. The interest in these topics is constantly changing over time. Their frequency occurrence diagrams can be used for effective prediction by the most straightforward simplification. In this paper, we use these diagrams to produce simple and human-readable rules that are able to predict the future trends of the most important keywords in 5 data sets of different types. A thorough analysis of the necessary input variables and parameters and their relation to the success rate is presented, as well.

KW - Prediction

KW - Time dynamics

KW - Topic evolution

KW - Topic similarities

UR - http://www.scopus.com/inward/record.url?scp=85047128697&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85047128697&partnerID=8YFLogxK

U2 - 10.1016/j.physa.2018.05.055

DO - 10.1016/j.physa.2018.05.055

M3 - Article

VL - 507

SP - 110

EP - 122

JO - Physica A: Statistical Mechanics and its Applications

JF - Physica A: Statistical Mechanics and its Applications

SN - 0378-4371

ER -