Stock market investing is undoubtedly challenging. Investors have to deal with random, vague and ambiguity stock price volatility before embarking on investment decision. Due to these weaknesses, the conventional model has several limitations; as a result investors are demanding for a new robust model which is able to represent their real situation to solve the uncertainty issues. In this study we developed a new fuzzy portfolio selection model using semi-variance as a risk measure integrates with investor’s judgment on assets’ future performance. Linear programming approach was used to optimize the portfolio risk and return. Empirical data showed that the model were able to derive a resourceful portfolio compared to the naïve portfolio selection.