Time series data mining for demand side decision support

Sengupta, Neha; Aloka, S.; Balakrishnan, N.; Ismail, Hamidah; Mathew, Satyajith

doi:10.1109/isgt-asia.2013.6698735

Cited by 5 publications

(4 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [10], Sengupta et al (2013) researches the convenience of time arrangement grouping systems to diminish the computational multifaceted nature of savvy framework improvement issues. The strategies center around the request side issue of neighborhood stockpiling measuring for sustainable combination, while featuring the significance and general materialness of these methods.…”

Section: Related Workmentioning

confidence: 99%

Efficient time series data classification using sliding window technique based improved association rule mining with enhanced support vector machine

Senthil¹,

Suseendran²

2018

IJET

View full text Add to dashboard Cite

Time series analysis is an important and complex problem in machine learning and statistics. In the existing system, Support Vector Machine (SVM) and Association Rule Mining (ARM) is introduced to implement the time series data. However it has issues with lower accuracy and higher time complexity. Also it has issue with optimal rules discovery and segmentation on time series data. To avoid the above mentioned issues, in the proposed research Sliding Window Technique based Improved ARM with Enhanced SVM (SWT-IARM with ESVM) is proposed. In the proposed system, the preprocessing is performed using Modified K-Means Clustering (MKMC). The indexing process is done by using R-tree which is used to provide faster results. Segmentation is performed by using SWT and it reduces the cost complexity by optimal segments. Then IARM is applied on efficient rule discovery process by generating the most frequent rules. By using ESVM classification approach, the rules are classified more accurately.

show abstract

Section: Related Workmentioning

confidence: 99%

Efficient time series data classification using sliding window technique based improved association rule mining with enhanced support vector machine

Senthil¹,

Suseendran²

2018

IJET

View full text Add to dashboard Cite

show abstract

“…For example, Fig. 1(a) shows the data chain with 70 sampling points, we can cut it into time intervals with 5 sampling points, i.e., [1,2,3,4,5], [6,7,8,9,10], etc. In each interval, we calculate the mean of total 5 data and get the mean to replace the former five points, which is shown by the "star" in the Fig.1.…”

Section: Introductionmentioning

confidence: 99%

Compressing sampling for time series big data

Bai-qi

Jin

2015

2015 34th Chinese Control Conference (CCC)

View full text Add to dashboard Cite

Due to the noise and uncertainty, it is necessary for time series big data to capture the key information with estimation methods. The Kalman filter with adaptive method by part of samples can give the high dimensional characteristics, reduce the computing cost and data uncertainty, but encounter the irregular estimation. The number of sample and the performance of the abstracted information have a tradeoff, which means we can use a suitable number of sample to abstract the key information of the series data. This paper discusses how to find the suitable sampling points for the time series data and the simulations show that the key information of time series big data can be extracted effectively with the compression amount number of sample data.

show abstract

“…For example, the data chain with 70 sampling points shown in Fig. 118 1(a), we can cut dawn it into the time interval with 5 sampling points, i.e., [1,2,3,4], [6,7,8,9,10,11], etc. In each interval, we calculate the mean of total 5 data and get the mean to replace the former five points, which is shown by the "star" in the Figure 1.To show clearly, Figure 1 (b) gives one of the time intervals from 26 to 30.…”

Section: Introductionmentioning

confidence: 99%

“…"Too fast" means that not only the data is big, but it must be processed quickly, for example, to perform fraud detection at a point of sale or decide which ad to show to a user on a webpage. In other words, most big data is with time series relationship [3,4]. "Too hard" is a catchall for data that doesn't fit neatly into one of existing processing tools or that needs some kind of analysis existing tools can't readily provide.…”

Section: Introductionmentioning

confidence: 99%