TY - GEN
T1 - High dimensional big data and pattern analysis
T2 - 2nd International Conference on Big Data Analytics, BDA 2013
AU - Lakshminarayan, Choudur K.
PY - 2013
Y1 - 2013
N2 - Sensors and actuators embedded in physical objects being linked through wired/wireless networks known as "internet of things" are churning out huge volumes of data (McKinsey Quarterly report, 2010). This phenomenon has led to the archiving of mammoth amounts of data from scientific simulations in the physical sciences and bioinformatics, to social media and a plethora of other areas. It is predicted that over 30 billion devices with 200 billion intermittent connections will be connected by 2020. The creation and archival of the massive amounts of data spawned a multitude of industries. Data management and up-stream analytics is aided by data compression and dimensionality reduction. This review paper will focus on some foundational methods of dimensionality reduction by examining in extensive detail some of the main algorithms, and points the reader to emerging next generation methods that seek to identify structure within high dimensional data not captured by 2 nd order statistics.
AB - Sensors and actuators embedded in physical objects being linked through wired/wireless networks known as "internet of things" are churning out huge volumes of data (McKinsey Quarterly report, 2010). This phenomenon has led to the archiving of mammoth amounts of data from scientific simulations in the physical sciences and bioinformatics, to social media and a plethora of other areas. It is predicted that over 30 billion devices with 200 billion intermittent connections will be connected by 2020. The creation and archival of the massive amounts of data spawned a multitude of industries. Data management and up-stream analytics is aided by data compression and dimensionality reduction. This review paper will focus on some foundational methods of dimensionality reduction by examining in extensive detail some of the main algorithms, and points the reader to emerging next generation methods that seek to identify structure within high dimensional data not captured by 2 nd order statistics.
KW - Canonical correlation Analysis
KW - Dimensionality Reduction
KW - Exploratory Projection Pursuit
KW - Factor Analysis
KW - Independent Component Analysis
KW - Multivariate Analysis
KW - Principal Component Analysis
KW - Projections
UR - http://www.scopus.com/inward/record.url?scp=84893342017&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84893342017&partnerID=8YFLogxK
U2 - 10.1007/978-3-319-03689-2_5
DO - 10.1007/978-3-319-03689-2_5
M3 - Conference contribution
AN - SCOPUS:84893342017
SN - 9783319036885
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 68
EP - 85
BT - Big Data Analytics - Second International Conference, BDA 2013, Proceedings
Y2 - 16 December 2013 through 18 December 2013
ER -