000 02830nam a2200385Ia 4500
001 76158
003 ft6075
005 20251124122034.0
008 190206n 000 0 eng d
022 _a2
040 _erda
041 _aengtag
050 _aQA76.75 D47 2016
082 _a.
100 _aErica Joy P. Dequito and Krizsha Rei P. Enero.
245 0 _aA further enhancement of the closet algorithm applied in market basket analysis
264 _a.
_b.
_cc2016
300 _bUndergraduate Thesis: (BSCS major in Computer Science) - Pamantasan ng Lungsod ng Maynila, 2016.
336 _b.
_atext
_2rdacontent
337 _b.
_aunmediated
_2rdamedia
338 _30
_b.
_avolume
_2rdacarrier
385 _a2
506 _a5
520 _aABSTRACT: Data mining is the practice of automatically searching large stores of data to discover patterns and trends that go beyond simple analysis. Data mining uses sophisticated mathematical algorithms to segment the data and evaluate the probability of future events. Data mining is also known as Knowledge Discovery in Data. One of the best and well-known algorithms under data mining field is the Closet Algorithm developed by Jain Pei, Jiawei Han and Runying Mao. It is mainly remarkable for its efficiency for producing association rules. This study discusses Closet Algorithm and its shortcomings such as not having an optimal solution when the frequencies of distribution to all transactions that are equal, it doesn't take consideration for the confidence threshold for mining the most effective association rules and lastly, it doesn't consider the positioning of items in the application. With these problems, the researchers have come up with solutions. First they considered the quantity of items, then generated the minimum confidence threshold and lastly they generated the positioning of items in the application. With these problems, the researchers have come up with solutions. First, they considered the quantity of items, then they generated the minimum confidence threshold and lastly they generated the positioning of items in the algorithm itself. After various tests and simulations, the algorithm is now enhanced and now has an optimal solution when the frequencies of distribution to all transactions are equal. And the enhanced algorithm also has the capability to decide on the best possible location and promotion of items. Also, the existing algorithm does not have a confidence threshold. With the enhanced algorithm, it can now generate the confidence percentage for mining the most strong and effective association rules.
526 _aF
540 _a5
600 _c2
_x0
610 _t2
_k0
_d6
611 _n2
_c0
630 _p20
650 _y20
651 _y20
655 _y2
942 _alcc
_cBK
_2ddc
999 _c25298
_d25298