TY - GEN
T1 - Identifying Non-pulsar Radiation and Predicting Chess Endgame Result Using ARSkNN
AU - Agarwal, Yash
AU - Kumar, Ashish
AU - Bhatnagar, Roheet
AU - Srivastava, Sumit
PY - 2020/1/1
Y1 - 2020/1/1
N2 - We are currently living in a data age. Due to the expansion of Internet of Things platform, there has been an upsurge in the number of devices connected to the Internet. Every device, ranging from smart sensors and smart phones to systems installed in manufacturing units, hospitals and vehicles is generating data. Such developments have not only escalated the generation of data but also created a need for analysis of raw data to identify patterns. Thus, data mining techniques are being deployed extensively to extract information. The accuracy and effectiveness of data mining techniques in providing better outcomes and cost-effective methods in various domains has already been established. Usually, in supervised learning, distance estimation is used by instance-based learning classifiers like kNN. In this analysis, the regular kNN classifier has been compared with ARSkNN which instead of following the conventional procedure of distance estimation uses the mass estimation approach. ARSkNN has been proved to be commensurate (or superior) to kNN in accuracy and has been found to reduce the computation time drastically on datasets chosen for this analysis.
AB - We are currently living in a data age. Due to the expansion of Internet of Things platform, there has been an upsurge in the number of devices connected to the Internet. Every device, ranging from smart sensors and smart phones to systems installed in manufacturing units, hospitals and vehicles is generating data. Such developments have not only escalated the generation of data but also created a need for analysis of raw data to identify patterns. Thus, data mining techniques are being deployed extensively to extract information. The accuracy and effectiveness of data mining techniques in providing better outcomes and cost-effective methods in various domains has already been established. Usually, in supervised learning, distance estimation is used by instance-based learning classifiers like kNN. In this analysis, the regular kNN classifier has been compared with ARSkNN which instead of following the conventional procedure of distance estimation uses the mass estimation approach. ARSkNN has been proved to be commensurate (or superior) to kNN in accuracy and has been found to reduce the computation time drastically on datasets chosen for this analysis.
UR - http://www.scopus.com/inward/record.url?scp=85076461769&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85076461769&partnerID=8YFLogxK
U2 - 10.1007/978-981-13-8406-6_65
DO - 10.1007/978-981-13-8406-6_65
M3 - Conference contribution
AN - SCOPUS:85076461769
SN - 9789811384059
T3 - Smart Innovation, Systems and Technologies
SP - 691
EP - 700
BT - Smart Systems and IoT
A2 - Somani, Arun K.
A2 - Shekhawat, Rajveer Singh
A2 - Verma, Vivek Kumar
A2 - Mundra, Ankit
A2 - Srivastava, Sumit
PB - Springer Paris
T2 - 2nd International Conference on Smart IoT Systems: Innovations and Computing, SSIC 2019
Y2 - 18 January 2019 through 20 January 2019
ER -