scholarly journals TEXT MINING TRANSPORTATION RESEARCH GRANT BIG DATA: KNOWLEDGE EXTRACTION AND PREDICTIVE MODELING USING FAST NEURAL NETS

2017 ◽  
Vol 7 (3) ◽  
Keyword(s):  
Big Data ◽  
Text Mining ◽  
Predictive Modeling ◽  
Knowledge Extraction ◽  
Neural Nets ◽  
Transportation Research ◽  
Research Grant

scholarly journals Big Data and Digital Aesthetic, Arts, and Cultural Education: Hot Spots of Current Quantitative Research

2019 ◽  
pp. 089443931988845 ◽  
Author(s):  
Alexander Christ ◽  
Marcus Penthin ◽  
Stephan Kröner
Keyword(s):  
Big Data ◽  
Text Mining ◽  
Predictive Modeling ◽  
Hot Spots ◽  
Quantitative Research ◽  
Hot Spot ◽  
Future Research ◽  
Original Research ◽  
Cultural Education ◽  
Data Volume

Systematic reviews are the method of choice to synthesize research evidence. To identify main topics (so-called hot spots) relevant to large corpora of original publications in need of a synthesis, one must address the “three Vs” of big data (volume, velocity, and variety), especially in loosely defined or fragmented disciplines. For this purpose, text mining and predictive modeling are very helpful. Thus, we applied these methods to a compilation of documents related to digitalization in aesthetic, arts, and cultural education, as a prototypical, loosely defined, fragmented discipline, and particularly to quantitative research within it (QRD-ACE). By broadly querying the abstract and citation database Scopus with terms indicative of QRD-ACE, we identified a corpus of N = 55,553 publications for the years 2013–2017. As the result of an iterative approach of text mining, priority screening, and predictive modeling, we identified n = 8,304 potentially relevant publications of which n = 1,666 were included after priority screening. Analysis of the subject distribution of the included publications revealed video games as a first hot spot of QRD-ACE. Topic modeling resulted in aesthetics and cultural activities on social media as a second hot spot, related to 4 of k = 8 identified topics. This way, we were able to identify current hot spots of QRD-ACE by screening less than 15% of the corpus. We discuss implications for harnessing text mining, predictive modeling, and priority screening in future research syntheses and avenues for future original research on QRD-ACE.

scholarly journals Big Data: New Tricks for Econometrics

2014 ◽  
Vol 28 (2) ◽  
pp. 3-28 ◽  
Author(s):  
Hal R. Varian
Keyword(s):  
Big Data ◽  
Linear Models ◽  
Neural Nets ◽  
Machine Learning Techniques ◽  
Support Vector ◽  
Learning Techniques ◽  
Model Complex ◽  
Vector Machines ◽  
Sheer Size ◽  
Complex Relationships

Computers are now involved in many economic transactions and can capture data associated with these transactions, which can then be manipulated and analyzed. Conventional statistical and econometric techniques such as regression often work well, but there are issues unique to big datasets that may require different tools. First, the sheer size of the data involved may require more powerful data manipulation tools. Second, we may have more potential predictors than appropriate for estimation, so we need to do some kind of variable selection. Third, large datasets may allow for more flexible relationships than simple linear models. Machine learning techniques such as decision trees, support vector machines, neural nets, deep learning, and so on may allow for more effective ways to model complex relationships. In this essay, I will describe a few of these tools for manipulating and analyzing big data. I believe that these methods have a lot to offer and should be more widely known and used by economists.

Sentiment Analysis with Text Mining in Contexts of Big Data

2017 ◽  
Vol 13 (3) ◽  
pp. 47-67 ◽  
Author(s):  
Carina Sofia Andrade ◽  
Maribel Yasmina Santos
Keyword(s):  
Big Data ◽  
Text Mining ◽  
Sentiment Analysis ◽  
Decision Making Process ◽  
Practical Applications ◽  
Information Models ◽  
Evolution Of Technology ◽  
The Common ◽  
Analyse Data

The evolution of technology, along with the common use of different devices connected to the Internet, provides a vast growth in the volume and variety of data that are daily generated at high velocity, phenomenon commonly denominated as Big Data. Related with this, several Text Mining techniques make possible the extraction of useful insights from that data, benefiting the decision-making process across multiple areas, using the information, models, patterns or tendencies that these techniques are able to identify. With Sentiment Analysis, it is possible to understand which sentiments and opinions are implicit in this data. This paper proposes an architecture for Sentiment Analysis that uses data from the Twitter, which is able to collect, store, process and analyse data on a real-time fashion. To demonstrate its utility, practical applications are developed using real world examples where Sentiment Analysis brings benefits when applied. With the presented demonstration case, it is possible to verify the role of each used technology and the techniques adopted for Sentiment Analysis.

Sign in / Sign up

Export Citation Format

Share Document