A Trend Analysis of Machine Learning Research with Topic Models and Mann-Kendall Test

Abstract

This paper aims to systematically examine the literature of machine learning for the period of 1968~2017 to identify and analyze the research trends. A list of journals from well-established publishers ScienceDirect, Springer, JMLR, IEEE (approximately 23,365 journal articles) related to machine learning is used to prepare a content collection. To the best of our information, it is the first effort to comprehend the trend analysis in machine learning research with topic models: Latent Semantic Analysis (LSA), Latent Dirichlet Allocation (LDA), and LDA with Coherent Model (LDA_CM). The LDA_CM topic model gives the highest topic coherence amongst all topic models under consideration. This study provides a scientific ground that helps to overcome the subjectivity of collective opinion. The Mann-Kendall test is used to understand the trend of the topics. Our findings provide indicative of paradigmatic shifts in research methodology of significant patterns of topical prominence and the evolving research areas. It is used to highlight the evolution regarding the previous and recent trends in research topics in the area of machine learning. Understanding such an intellectual structure and future trends will assist the researchers to adopt the divergent developments of this research in one place. This paper analyzes the overall trends of the machine learning research since 1968, based on the latent topics identified in the period of 2007~2017 that may be helpful to the researchers exploring the recommended areas and publish their research articles.

Exploring Latent Topics and International Research Trends in Competency-Based Education Using Topic Modeling

“…It provides a theoretical background that can help understand the document creation mechanism, and documents can be automatically organized and summarized. Topic models have been used to analyze research trends in various fields, such as education [24,25], statistics [26], machine learning [27], biochemistry [28], and manufacturing [29].…”

“…Using Genism's LDA Model, LDA analysis was conducted. To find the optimal hyper parameter values, the setting as shown in Table 2 was used by referring to related papers [27,34]. To determine K, the number of topics was varied from 10 to 30, and the coherence and perplexity values were compared.…”

Discovering Trends of Mobile Learning Research Using Topic Modelling Approach

“…Amado, et al [23] presented a research literature analysis on big data in marketing by applying the text mining approach. Sharma, et al [24] discovered and analyzed the research trends in machine learning by investigating the literature of machine learning from 1968-2017. They employed several methods to identify the topics, such as LSA (Latent Semantic Analysis), LDA (Latent Dirichlet Allocation), and LDA_CM (Latent Dirichlet Allocation with Coherence Model).…”

Evaluating Research Trends from Journal Paper Metadata, Considering the Research Publication Latency

“…Due to their reduced sensitivities to outliers [2], the lack of assumptions concerning the data sample distribution [3] or homoscedasticity [4], nonparametric trend tests tend to be favored by researchers over parametric methods. In particular, the Mann-Kendall (MK) test statistic being a robust trend indicator when dealing with censored data, arbitrary non-Gaussian data distributions or time series with missing observations [5] have become almost standard methods for NLP applications [1,[6][7][8][9][10][11][12].…”

“…In medicine-related NLP applications, Marchini et al [6] used the same method to analyze the urologic research trends described by 12 key terms, Chakravorti et al [13] employed the MK trend evaluation method to detect and characterize mental health trends in online discussions, while Modave et al [14] evaluated perception and attitude trends in breast cancer twitter messages. Moreover, Sharma et al [7] used the Mann-Kendall test to understand machine learning research topic trends using metadata from journal papers. In [8], Zou analyzed the journal titles and abstracts to explore the temporal popularity of 50 drug safety research trends over time using the MK test, while Neresini et al [15] used the Hamed and Rao [2] MK variant to extract trends from correlated time series.…”