### Table 1. Parameters of Evolutionary and Apriori algorithms.

2003

Cited by 5

### Table 1: Set of URLs obtained by Apriori algorithm

2001

"... In PAGE 5: ... Table1 shows that most of the item sets are sub- sets of other item sets. Only a small number of large clusters are extracted as a solution.... ..."

Cited by 4

### Table 1 Apriori-TFPC algorithm

"... In PAGE 5: ... The nature of the rules produced is there- fore very much dependent on the threshold value used. A high level view of the Apriori-TFP CR generation al- gorithm is presented in Table1 where a4 a1a0 indicates the set of level a5 T-tree nodes and a9 a4 a0 the set of candidate level a5 T-tree nodes. The constants a44a47a46 a1 a5a22a6a34a8 and a44a60a46 a1 a16a18a17 a1a20a19 in- dicate the minimum support and confidence thresholds sup- plied by the user.... In PAGE 5: ... Remember that although the algorithm compares a calculated confidence value to a threshold value when deciding whether to add a rule to a2 or not, this could equally well be a Laplace or WRA accuracy value or a a2 a3 value. To prevent the T-tree from growing too large an addi- tional test (not shone in Table1 ) is included in the gener- ation process in that if the number of nodes in the T-tree is greater than 80,000; on completion of a level, no further levels are generated. On completion of the generation process a default rule is also identified; this is the class associated with the last rule in a2 .... ..."

### TABLE II. THE RUNTIME COMPARISON OF OUR ALGORITHM AND APRIORI-BASED ALGORITHM

2003

Cited by 3

### Table 2. Numbers of candidates analyzed by the Apriori-like algorithm

"... In PAGE 8: ... Table2 presents numbers of candidates analyzed by the Apriori-like algorithm for different numbers of event attributes (denoted as NATR) and their domain sizes ... ..."

### Table 1. Sample Web Transactions and Frequent Item- sets generated by Apriori algorithm

2003

"... In PAGE 4: ...Itiseasy to observe that in this algorithm the search process re- quires only O(|w|) time given active session window w. To illustrate the process, consider the example transaction set given in Table1 (top). Using these transactions, the Apriori algorithm with a frequency threshold of 4 (minimum support of 0.... In PAGE 4: ... Figure 1 shows the Frequent Itemsets Graph constructed based on the ABCE (4) ABC (4) ABE (5) ACE (4) BCE (4) AB (5) AC (4) AE (5) BE (5) BC (4) CE (4) A (5) B (6) C (4) E (5) O Depth 0 Depth 1 Depth 2 Depth 3 Depth 4 Figure 1. The Frequent itemsets Graph for example frequent itemsets in Table1 . Now, given user active session window lt;B,E gt;, the recommendation gener- ation algorithm finds items A and C as candidate rec- ommendations.... ..."

Cited by 20

### Table 1: Results of the top-down, naieve A-priori algorithm on randomized Bernoulli Databases.

"... In PAGE 30: ...c A randomized Bernoulli database (having no repeated tuples) is generated of the appro- priate width and length, and then the top-down naieve key- nding function is applied. Table1 shows the results obtained. Note that we use an e cient key-checking algorithm based on a version of quicksort.... In PAGE 30: ...ased on a version of quicksort. The algorithm is given in gure 5. Input: X = fA1; :::; Akg, a set of attributes; r = ft1; :::; tmg a database instance Output: TRUE if X is a key for r, FALSE otherwise if (jrj gt; 2jXj) return FALSE; QuickSort(r, X); // Sort r on attributes in X i = 1; j = 2; while (i lt;= m ? 1 and j lt;= m) if ti = tj return FALSE; // duplicates will be next-to-next return TRUE; Figure 5: Key-checking predicate: determines if the set X is a key according to the database instance r. Table1 shows results for the worst, average, and best cases of the top-down naieve algorithm, respectively. Note that the worst case occurs when there is only one tuple in the database, since in this case every subset of size 1 is a minimal key.... ..."

### Table 1. A three-dimensional classification view of prevailing Apriori-like algorithms

"... In PAGE 2: ...earch direction: top-down vs. bottom-up. Although the first two aspects have been addressed in [7][8], no comparison has revealed the influence of the last aspect. Table1 shows a three-dimensional classification of prevailing Apriori-like algorithms. Table 1.... ..."