Please use this identifier to cite or link to this item:
https://ptsldigital.ukm.my/jspui/handle/123456789/475557Full metadata record
| DC Field | Value | Language |
|---|---|---|
| dc.contributor.advisor | Zalinda Othman, Dr. | - |
| dc.contributor.author | Huda Fathi Eshames (P50122) | - |
| dc.date.accessioned | 2023-10-05T06:39:57Z | - |
| dc.date.available | 2023-10-05T06:39:57Z | - |
| dc.date.issued | 2011-07-07 | - |
| dc.identifier.other | ukmvital:74393 | - |
| dc.identifier.uri | https://ptsldigital.ukm.my/jspui/handle/123456789/475557 | - |
| dc.description | Any abnormal patterns shown in Statistical Process Control (SPC) charts imply the presence of possible assignable causes and variances that may leads to the process performance deterioration. Therefore, timely detection and recognizer of patterns in control chart is very important in the SPC implementation. This research presents the classification techniques for anomaly detection in control chart patterns. The control chart dataset has its specific features that need specific data preprocessing procedures. It is crucial and involves a number of stages of data preparation procedures. Firstly, the Principle Component Analysis (PCA) is employed for attributes reduction. Secondly, the Piecewise Aggregate Approximation (PAA) and Symbolic Aggregate Approximation (SAX) are used as data representation. The preprocessed data are fitted to the classification algorithms in extracting important knowledge .The aim of this research is to present and investigate the performance of different classification methods for a set of large data in recognition of control chart patterns. The algorithms or methods tested are decision tree, support vector machine, MLP networks, JRiP algorithm and Ridor algorithm. A dataset obtained from UCI KDD (Synthetic Control Chart Time Series dataset) is used as a case study. The control chart data has dimensions of 600 rows and 60 columns with 6 different classes. The data will be used to test and justify the differences between the classification methods. The results obtained shown the feasibility of the data preprocessing approaches in reducing attributes/ dimensions. Therefore, in order to improve the efficiency and accuracy of mining task on high dimensional data. The simulation experimental results show that data mining and machine learning can be extremely beneficial in generates quality knowledge and may benefit the manufacturers. The results indicate that JRiP algorithm is the best algorithm and achieved highest detection accuracy about 99.66 % and lowest errors rate is 2.9871.,Master/Sarjana | - |
| dc.language.iso | eng | - |
| dc.publisher | UKM, Bangi | - |
| dc.relation | Faculty of Information Science and Technology / Fakulti Teknologi dan Sains Maklumat | - |
| dc.rights | UKM | - |
| dc.subject | Abnormal patterns | - |
| dc.subject | Control charts | - |
| dc.subject | Classification techniques | - |
| dc.subject | Process control -- Statistical methods -- Computer programs | - |
| dc.title | Abnormal patterns detection in control charts using classification techniques | - |
| dc.type | Theses | - |
| dc.format.pages | 107 | - |
| dc.identifier.callno | TS156.8.E837 2011 3 | - |
| dc.identifier.barcode | 000813 | - |
| Appears in Collections: | Faculty of Information Science and Technology / Fakulti Teknologi dan Sains Maklumat | |
Files in This Item:
There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.