Combining informetrics and trend analysis to understand past and current directions in electronic design automation

Abstract

There has been increasing interest in the study of research communities with the goal of optimizing their outcomes and impact. While current methods can predict future trends, they offer little insight about the causes of the trends. However, causal insight is important for strategic decision making to improve a community. This paper presents a new method to predict the possible causes for inefficiencies in a community by relating them to disconnections between trends, like trends in the number of publications, patents, citations, and so on. The method combines traditional scientometric and webometric metrics and metric predictions with a recent model for trend analysis in a community. The proposed method was used to analyze electronic design automation (EDA) domain. The analysis showed intriguing disconnections between the trends of the number of papers, number of granted patents, and impact of its main publications. The analysis suggests a slightly decreasing impact and visibility of EDA, while having less novel, commonly-accepted knowledge in the area. The gained insight suggests three possible strategic decisions to improve EDA community: avoiding to ignore new ideas, reducing the complexity of framed problems, and keeping a minimal gap between real-life needs and academic solutions.

Identifying the Technology Opportunities and the Technology Taxonomy for Railway Static Inverters With Patent Data Analytics

“…Time series analysis is a method of analyzing trends for future changes based on time series data recorded over time according to specific phenomena from the past to the present. It is different from other analyses in that it considers the flow of time [61], [62]. For patent data belonging to each technology topic, we converted it into time series data consisting of independent variables indicating the time point and dependent variables indicating the frequency to predict future trends [63], [64].…”

Identifying Hot Information Security Topics Using LDA and Multivariate Mann-Kendall Test

“…Its role becomes extremely important anytime researchers are searching for a new scientific niche to be explored and generally relies on systematic trend analysis in citation or key term time series. As a result, various trend analysis approaches designed to track topic evolution [15], to predict their future trends [16] or to discover hot research topics [3] have been proposed.…”