Topic model visualization with IPython

Karpovich, S. N.; Smirnov, Alexander; Teslya, Nikolay; Grigor'ev, A. Yu.

doi:10.23919/fruct.2017.8071303

Cited by 12 publications

(6 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Technically, to find the topics, we used the LDA class provided by the scikit-learn (Pedregosa et al, 2011) Python library, a popular scientific package for data analysis and machine learning. The found topics were visualized using pyLDAvis (Karpovich, Smirnov, Teslya, and Grigorev, 2017), a package for visualizing topics on iPython notebook. The visualization is built on the classical multi-dimensional scaling (MDS) (Cox and Cox, 2000), which assigns topic circles coordinates in the two-dimensional space such that their distances in the high-dimensional space are preserved.…”

Section: Proposed Mixed-methods Approachmentioning

confidence: 99%

Twitter data analysis for studying communities of practice in the media industry

Komorowski

Huu

Deligiannis

2018

Telematics and Informatics

View full text Add to dashboard Cite

Today, more and more physical communities of practice, a concept that describes a group of people that share a passion and interact regularly at events to exchange knowledge, utilize social media, such as Twitter. Brotaru, for instance, is such a physical community of practice for media professionals in Brussels. It is a monthly meet-up of videogame developers in various locations in Brussels. Furthermore, Twitter becomes widely acknowledged as important instrument for learning and community formation in the virtual world. But, do these communities of practice use Twitter only to promote their physical activities of learning? Or, are the activities of the physical communities further extended into the virtual world meaning that virtual communities of practice emerge from them? This article suggests a novel mixed-methods approach based on qualitative and quantitative data to measure the role of Twitter for physical communities of practice. The method applies different statistical measures and analysis on harvested Twitter data and additionally brings two of the most used methods in Twitter analysis together, social network analysis and text data analysis (a.k.a., content analysis). Four different communities of practice in Brussels' media industry and their activities and followers on Twitter have been analysed. The findings showed that the activities of the communities of practice extend into the Twitter sphere as the online communities are characterised by (1) a shared domain, (2) a lively community and (3) shared practices. The analysis further revealed that Twitter offers three main opportunities for the activities of communities of practice: it offers geographical extension; it gives temporal autonomy; and, it can be used to diversify the practices.

show abstract

Section: Proposed Mixed-methods Approachmentioning

confidence: 99%

Twitter data analysis for studying communities of practice in the media industry

Komorowski

Huu

Deligiannis

2018

Telematics and Informatics

View full text Add to dashboard Cite

show abstract

“…Also, the closeness of the topics can be interpreted as their thematic relevancy. In fact, in this visualization, those topics in which the common terms are more frequent have been located closer together (Karpovich et al , 2017; Sievert and Shirley, 2014).…”

Section: Research Resultsmentioning

confidence: 85%

A text mining-based review of the literature on dynamic capabilities perspective in information systems research

Talafidaryani

2020

MRR

View full text Add to dashboard Cite

Purpose While the dynamic capabilities perspective is the most cited strategic theory in the information systems field of research, little effort has been made to review and integrate the associate literature of this perspective in the field. Accordingly, this paper aims to systematically analyze the information systems literature on dynamic capabilities and provide a holistic understanding of the topical composition and trend of dynamic capabilities studies in information systems research. Design/methodology/approach Using latent Dirichlet allocation as the text analysis algorithm, the author conducted a topic modeling of the dynamic capabilities corpus in the information systems field of research to quantitatively review, summarize and classify the prior literature. The review covered 191 articles published on dynamic capabilities between 1998 and 2018 in pioneering information systems journals and conference proceedings. Findings In accordance with the topic modeling results, the topical composition of the dynamic capabilities corpus in information systems research dominantly includes seven themes titled T1. Information systems value, T2. Information systems change, T3. Digitalization, T4. Information systems agility, T5. Big data, T6. Information systems innovation and T7. Information systems alignment. Also, the overall and topical trend of dynamic capabilities studies in the information systems field of research were revealed. The trends indicated that the investigated domain and its prominent sub-domains have generally had positive productivity over the past years. Originality/value The current study contributes to the domain by developing knowledge and improving literature on dynamic capabilities in information systems research, discovering the main topics of interest for information systems researchers to deploying the dynamic capabilities perspective in their studies, and prioritizing the future information systems research on dynamic capabilities based on the identified trends of topics.

show abstract

“…State space models are used to represent the topics, and variational approximations are developed to carry out approximate posterior inference over the latent topics [22]. DTM in the Python Gensim package [23] was implemented to generate topics in different time slices, and the method mentioned in Karpovich et al [24] was reused to visualize topic changes.…”

Section: Automatic Content Analysismentioning

confidence: 99%

Demystifying COVID-19 publications: institutions, journals, concepts, and topics

Chen

Nguyen

2021

jmla

View full text Add to dashboard Cite

Objective: We analyzed the COVID-19 Open Research Dataset (CORD-19) to understand leading research institutions, collaborations among institutions, major publication venues, key research concepts, and topics covered by pandemic-related research.Methods: We conducted a descriptive analysis of authors’ institutions and relationships, automatic content extraction of key words and phrases from titles and abstracts, and topic modeling and evolution. Data visualization techniques were applied to present the results of the analysis.Results: We found that leading research institutions on COVID-19 included the Chinese Academy of Sciences, the US National Institutes of Health, and the University of California. Research studies mostly involved collaboration among different institutions at national and international levels. In addition to bioRxiv, major publication venues included journals such as The BMJ, PLOS One, Journal of Virology, and The Lancet. Key research concepts included the coronavirus, acute respiratory impairments, health care, and social distancing. The ten most popular topics were identified through topic modeling and included human metapneumovirus and livestock, clinical outcomes of severe patients, and risk factors for higher mortality rate.Conclusion: Data analytics is a powerful approach for quickly processing and understanding large-scale datasets like CORD-19. This approach could help medical librarians, researchers, and the public understand important characteristics of COVID-19 research and could be applied to the analysis of other large datasets.

show abstract

Topic model visualization with IPython

Cited by 12 publications

References 7 publications

Twitter data analysis for studying communities of practice in the media industry

Twitter data analysis for studying communities of practice in the media industry

A text mining-based review of the literature on dynamic capabilities perspective in information systems research

Demystifying COVID-19 publications: institutions, journals, concepts, and topics

Contact Info

Product

Resources

About