000			02582nam a2200277Ia 4500
001			76992
003			ft6161
005			20251105173938.0
008			190313n 000 0 eng d
040			_erda
041			_aengtag
050			_aQA76.75 D56 2011
082			_a.
100			_aDioquino, Rica I.
245		0	_aAn Enhancement of charm algorithm applied to shopping cart system
264			_a. _b. _cc2011
300			_bUndergraduate Thesis: (BSCS major in Computer Science) - Pamantasan ng Lungsod ng Maynila, 2011.
336			_b. _atext _2rdacontent
337			_30 _b. _aunmediated _2rdamedia
338			_30 _b. _avolume _2rdacarrier
505			_aABSTRACT: Recent years, man started to study extracting patterns from data. And as year pass; Data mining is one of the most common fields of study. Data mining is branch of computer science. It is a process of extracting patterns from a large datasets by combining methods statistics, artificial intelligence with database management. Charm algorithm is an efficient algorithm for closed association rule mining. Although the algorithm is efficient enough, some flaws are found by one proponent and these are the following: 1. The existing algorithm scanned frequently the database to be able to count the frequent support of each items. In this process there are rereading of items resulting of redundant scanning in database. 2. In the existing algorithm, the minimal support is set to 50%. If the user chose an item and it doesn’t reach the qualified minimal support, there will be no output to be produced. And since the item being tested doesn’t reach the minimal support, that item will not undergo the algorithm process that computes for the closed item set which refers to association of one item to another item. Even the user chose item that reach the minimum support of 50, the items with less than the minimum support will also be eliminated as initial node in searching process. This problem has been highlighted because one important part of the application is to promote items for better merchandizing technique. 3. The existing algorithm produced different combination of items according to their association with other items and their support in the database. Hence, there are redundant items being produced as an output. The existing algorithm lacks in checking this redundant items being produced. This problem is highlighted to be able to produce an output that is precise, organize and more easy to analyze.
526			_aF
540			_a5
655			_a.
942			_alcc _cMS _2lcc
999			_c10396 _d10396