Unveiling Topics from Scientific Literature on the Subject of Self-driving Cars using Latent Dirichlet Allocation

Ayele, Workneh Y.; Juell-Skielse, Gustaf

doi:10.1109/iemcon.2018.8615056

Cited by 5 publications

(10 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Scopus has better coverage of journals [55] and contains the latest and larger datasets of scholarly literature than Web of Science [56]. Exploring and understanding data can be carried out using a spreadsheet application [37,57].…”

Section: How Can the Activities In This Phase Be Done?mentioning

confidence: 99%

“…After the tokenization of each document, the lemmatization of terms was done to convert terms to their root forms. Instead of lemmatization, stemming could also be done, as illustrated in [37]. After tokenization, bigrams were computed and added to the tokens list.…”

Section: How Can the Activities In This Phase Be Done?mentioning

confidence: 99%

“…The inputs to the model generation process are the number of topics, preprocessed text corpus, and dictionary. Perplexity measures are used to determine the number of topics [37] or coherence scores [60]. The best fit model could be generated by comparing the semantic interpretability of models [62].…”

Section: How Can the Activities In This Phase Be Done?mentioning

confidence: 99%

“… Determine the optimum number of topics -it is possible to determine an optimum number of topics through perplexity [37] or coherence scores [60].…”

Section: How Can the Activities In This Phase Be Done?mentioning

confidence: 99%

“…Also, with regards to urban planning, efficient parking is a characteristic of self-driving vehicles [36]. However, there are growing concerns utilizing self-driving technology such as ethical issues [37], transportation system, affordability, safety, control, and liabilities [36], and the impact of self-driving technologies on urban planning [37].…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Adapting CRISP-DM for Idea Mining

Ayele¹

2020

IJACSA

Self Cite

View full text Add to dashboard Cite

Data mining project managers can benefit from using standard data mining process models. The benefits of using standard process models for data mining, such as the de facto and the most popular, Cross-Industry-Standard-Process model for Data Mining (CRISP-DM) are reduced cost and time. Also, standard models facilitate knowledge transfer, reuse of best practices, and minimize knowledge requirements. On the other hand, to unlock the potential of ever-growing textual data such as publications, patents, social media data, and documents of various forms, digital innovation is increasingly needed. Furthermore, the introduction of cutting-edge machine learning tools and techniques enable the elicitation of ideas. The processing of unstructured textual data to generate new and useful ideas is referred to as idea mining. Existing literature about idea mining merely overlooks the utilization of standard data mining process models. Therefore, the purpose of this paper is to propose a reusable model to generate ideas, CRISP-DM, for Idea Mining (CRISP-IM). The design and development of the CRISP-IM are done following the design science approach. The CRISP-IM facilitates idea generation, through the use of Dynamic Topic Modeling (DTM), unsupervised machine learning, and subsequent statistical analysis on a dataset of scholarly articles. The adapted CRISP-IM can be used to guide the process of identifying trends using scholarly literature datasets or temporally organized patent or any other textual dataset of any domain to elicit ideas. The ex-post evaluation of the CRISP-IM is left for future study.

show abstract

Section: How Can the Activities In This Phase Be Done?mentioning

confidence: 99%

Section: How Can the Activities In This Phase Be Done?mentioning

confidence: 99%

Section: How Can the Activities In This Phase Be Done?mentioning

confidence: 99%

“… Determine the optimum number of topics -it is possible to determine an optimum number of topics through perplexity [37] or coherence scores [60].…”

Section: How Can the Activities In This Phase Be Done?mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Adapting CRISP-DM for Idea Mining

Ayele¹

2020

IJACSA

Self Cite

View full text Add to dashboard Cite

show abstract

Identifying Emerging Trends and Temporal Patterns About Self-driving Cars in Scientific Literature

Ayele

Akram

2019

Advances in Intelligent Systems and Computing

Self Cite

View full text Add to dashboard Cite

Self-driving is an emerging technology which has several benefits such as improved quality of life, crash reductions, and fuel efficiency. There are however concerns regarding the utilization of self-driving technology such as affordability, safety, control, and liabilities. There is an increased effort in research centers, academia, and the industry to advance every sphere of science and technology yet it is getting harder to find innovative ideas. However, there is untapped potential to analyze the increasing research results using visual analytics, scientometrics, and machine learning. In this paper, we used scientific literature database, Scopus to collect relevant dataset and applied a visual analytics tool, CiteSpace, to conduct co-citation clustering, term burst detection, time series analysis to identify emerging trends, and analysis of global impacts and collaboration. Also, we applied unsupervised topic modeling, Latent Dirichlet Allocation (LDA) to identify hidden topics for gaining more insight about topics regarding self-driving technology. The results show emerging trends relevant to self-driving technology and global and regional collaboration between countries. Moreover, the result form the LDA shows that standard topic modeling reveals hidden topics without trend information. We believe that the result of this study indicates key technological areas and research domains which are the hot spots of the technology. For the future, we plan to include dynamic topic modeling to identify trends.

show abstract

Eliciting Evolving Topics, Trends and Foresight about Self-driving Cars Using Dynamic Topic Modeling

Ayele

Juell-Skielse

2020

Advances in Intelligent Systems and Computing

View full text Add to dashboard Cite

Unveiling Topics from Scientific Literature on the Subject of Self-driving Cars using Latent Dirichlet Allocation

Cited by 5 publications

References 17 publications

Adapting CRISP-DM for Idea Mining

Adapting CRISP-DM for Idea Mining

Identifying Emerging Trends and Temporal Patterns About Self-driving Cars in Scientific Literature

Eliciting Evolving Topics, Trends and Foresight about Self-driving Cars Using Dynamic Topic Modeling

Contact Info

Product

Resources

About