The agricultural industry is vital for a country's economic well-being, but crop disease is a major challenge. Peppercorn crops require special attention due to disease concerns. Data mining has been used for crop disease identification but lacks knowledge utilization. A study integrates data mining with knowledge-based systems for peppercorn crop disease diagnosis and treatment. Experiments were conducted using four algorithms named JRip, PART, J48, and REPTree on the peppers dataset. All experiments for each algorithm were conducted containing 9927 instances and four classes namely Fungus, Insects, Virus, and Bacteria disease types of the crop. Classification algorithms were used to develop a predictive model and rule-based knowledge representation was employed to diagnose and treat the crop. The system's performance was tested with domain experts and user acceptance with promising results of 90.5 and 86.8% respectively with an average Precision and Recall of 96 % and 97.2% overall performance result.