Negative and Positive Association Rules Mining from Text Using Frequent and Infrequent Itemsets

Algorithm 1

Input:  TD: Transaction (Document) Database; ms: minimum support threshold;

Output:  FIS: frequent itemsets; inFIS: infrequent itemsets;

(1) initialize FIS = Φ; inFIS = Φ;

(2) temp₁ = ∀𝐴∣𝐴∈ top(N) IDF terms;          /* get all 1-itemsets being in top-N IDF items */

(2.1) 𝐅𝐈𝐒_𝟏 = {𝐴∣𝐴∈ temp₁ and support (𝐴) ≥ (ms)};     /* get frequent 1-itemsets */

(2.2) 𝐢𝐧𝐅𝐈𝐒_𝟏 = temp₁ − FIS₁;             /* all infrequent 1-itemsets */

𝑘=2;                     /* initialize for itemsets greater than 1 */

(3) while (temp_𝑘−1≠ Φ) do begin

   (3.1) 𝐶_𝑘 = generate (temp_𝑘−1, ms);           /* candidate 𝑘-itemsets */

   (3.2) for each transaction 𝑡∈TD

      do begin               /* scan database TD*/

      𝐶_𝑡 = subset(𝐶_𝑘,𝑡);            /* get temp candidates in transaction 𝑡 */

      for each candidate 𝑐∈𝐶_𝑡

        c.count⁺⁺;             /* increase count of itemsets if it exists in transaction */

      end;

   (3.3) c.support = 𝑐.count|TD|;             /* calculate support of candidate 𝑘-itemset */

   (3.4) temp_𝑘={𝑐∣𝑐∈𝐶_𝑘 and (𝑐.support≥minsup)};  /* add to temp 𝑘-itemsets */

(4) 𝐅𝐈𝐒_𝐤 = {𝐴∣𝐴∈ temp_𝑘 and 𝐴. support ≥ ms)};        /* add to frequent 𝑘-itemsets if supp(𝐴) greater than minsupp*/

(5) 𝐢𝐧𝐅𝐈𝐒_𝐤 = temp_𝑘−FIS_𝑘;               /* add to in-frequent 𝑘-itemsets having support less than minsupp*/

(6) FIS = ∪_𝑘  FIS_𝑘;                   /* add generated frequent 𝑘-itemsets to FIS */

(7) inFIS = ∪_𝑘  inFIS_𝑘;               /* add generated in-frequent 𝑘-itemsets to inFIS */

(8) k ⁺⁺;                     /* increment itemset size by 1 */

end;

(9) return FIS and inFIS;