| Input: TextDocuments (T1, T2, T3, …, Tn) under different Category (C1, C2, C3, …, Cn) |
| Output: Classified-Textual-Features |
| | Step 1: Gather the text documents from 1 to n, such as T1, T2, …, Tn. |
| | Step 2: Accumulate the category of text documents such as C1, C2, C3, …, Cn. |
| | Step 3: Initiate the looping for each accumulated category of text documents. |
| | Step 4: Generate another looping structure for text documents from 1 to n. |
| | Step 5: Split the text document features separately. |
| | Step 6: Eliminate the StopWords. |
| | Step 7: Assign Frequency for text documents based on the category. |
| | Step 8: Generate Stemming Process based on the frequency generated in step 7. |
| | Step 9: Add the result to the data store and exit from the looping structures of step 3 and step 4. |
| | Pseudocode: |
| for (each text_Category [C1-to-Cn] w.r.t. [T1-to-Tn]) |
| for (each text_Documents [T1-to-Tn] w.r.t. [C1-to-Cn]) |
| Split the text_features and assign it to Fj |
| Eliminate the StopWords from the text_features and assign the resultant to SFj. |
| Assign the Fj + SFj to Tj |
| Generate the frequency for each Tj based on category (Cj). |
| Perform Stemming based on SFj & Cj. |
| Add the resultant features to Data store (DS). |
| end for |
| end for |
| | Step 10: Initiate the textual_population randomly. |
| | Step 11: Start the while loop and iterate up to maximum level of feature presented into the given text document. |
| | Step 12: Raise the iteration level by 1 until the loop ends. |
| | Step 13: Estimate the fitness function of every term frequency. |
| | Step 14: Select the fitness property for each iterative word over the loop. |
| | Step 15: Perform the crossover functionality. |
| | Step 16: Raise the text-population levels for every crossover iteration generated in step 15. |
| | Step 17: Apply the decision tree procedure with optimum value of confidence threshold. |
