Đang chuẩn bị liên kết để tải về tài liệu:
High-Performance Parallel Database Processing and Grid Databases- P11

Không đóng trình duyệt đến khi xuất hiện nút TẢI XUỐNG Tải xuống

High-Performance Parallel Database Processing and Grid Databases- P11: Parallel databases are database systems that are implemented on parallel computing platforms. Therefore, high-performance query processing focuses on query processing, including database queries and transactions, that makes use of parallelism techniques applied to an underlying parallel computing platform in order to achieve high performance. | 480 Chapter 17 Parallel Clustering and Classification Rec Weather Temperature Time Day Jog Target Class 1 Fine Mild Sunset Weekend Yes 2 Fine Hot Sunset Weekday Yes 3 Shower Mi Id Midfay WaeCdad No 4 Thunderstorm Cool Dawn Weekend No 5 Shower Hot Sunset Weekday Yes 6 Fine Hot Middea Weekday do 7 Fine Cool Dawn Weekend do 8 Thunderstorm Cool Middea Weekday No 9 Fine Cool Mlddao Weekday Yes 10 Fine Mild Middfo Weekday des 11 Shower Hot Dawn Weekend No 12 Shower Mild Dawn Weekday Yo 13 Fine Cool Dawn Weekday No 14 Thunderstorm Mild Sunset Weekend No 15 Thunderstorm Hot Middea Weekday No Figure 17.11 Training data set thunderstorm whereas the possible volum for temperaturearehot mild and cool. Continuous values are real numbers e.g. heights of a person in centimetres . Figure 17.11 showsthe training t tiaset for the decision tree shown previously. This training data set consists of only 15 records. For simplicity only categorical attributes are used in this example. Examining the first record and matching it with the decision tree inFiaere 17.10 the target isa Yer forfmewenmer eiKlimld temperature disregcdmgiheofteitwoalttribulfiTeis is becauseall reeordtin this training data set follow russule iserecordslaad 10f. Other records.riicli er records 9 and 13 usgcli die load attribdtes. 17.3.2 DecisionTeeeClacstficatihni Peocesses Decision Tree Algorishm There are many different algorithms to construct a decision tree such as ID3 C4.5 Spriiil etc. Consrnrnga decision Been jeeCyh enohe process. Ath d start all training records are at tlietootnode. Then il.phididonttlieOaining records recursively by choosing one attribute at a time. The process is repeated for the partitioned data set.Therea.irsnm stops when a stopping condition is reached which is when all og 116 1 -60 inihepsriilind 11301 1 0316 61-13 label. Figure 17.12 shows analgorlalmfoseoestruciingadeer1iogtree. Tfe decision tree constructie a di 5heilii -n - coi fim ti eil nd. Itcosseucir the tree using a .