AN INTELLIGENT MACHINE LEARNING-BASED TRADING SIGNAL PREDICTION SYSTEM FOR THE IQ OPTION OTC MARKET: EVIDENCE FROM INVESTMENT HOUSE, ADAMAWA STATE, NIGERIA

Abstract

Short-term financial markets, particularly Over-the-Counter (OTC) synthetic digital options, are characterized by high volatility, non-linear noise, and continuous operation. Traditional manual trading strategies in these environments suffer from human psychological vulnerabilities, decision latency, inconsistence technical analysis, and physical fatigue. This paper presents a design, implementation, and evaluation of An Intelligent Machine Learning-Based Trading Signal Prediction System for The IQ Option OTC currency pairs, using Investment House in Adamawa State, Nigeria-A localized operational case study. The aim of the study was to establish an automated, data-driven decision-support mechanism that replaces subjective manual chart interpretation. The methodology encompasses historical candle data retrieval from the IQ Option API, robust data cleaning, chronological data partitioning to prevent temporal leakage, and engineering an eighteen-variable technical feature space. A Random Forest classifier configured with 300 decision trees was trained to predict price movement directions classified into BUY, HOLD, and SELL targets. To minimize false positives, a strict confidence-based filtering module was integrated, requiring a minimum 70% class probability before a prediction is qualified as an actionable trading signal. Qualified signals are automatically logged into CSV storage and disseminated in real time via the Telegram Bot API. System evaluation confirmed successful real-time data acquisition, automated 18-feature extraction, consistent model execution, and low-latency notification delivery without executing automated orders, preserving full user execution control. The findings demonstrate that structured supervised group learning coupled with confidence thresholding provides a reliable framework for short-term financial decision support in localized FinTech contexts. Periodic model retraining and demo-environment validation are recommended prior to operational deployment.