Abstarct: Stock market forecasting using data mining methods is one of the most important topics in investment and financial market research. Many efforts have been made to predict this market using traditional methods. However, due to the increasing volume of information, these methods are not suitable for analyzing this amount of information. Data mining is able to discover hidden patterns and predict future trends in the stock market. The stock market can be considered a data mining problem and clustering is one of the data mining methods that is a good strategy to advance foresight and guidance of investors decisions. Stock liquidity means the ability to buy and sell stocks in the shortest time and at the lowest cost. Therefore, investors choose companies that have high liquidity. In this study, clustering of companies active in the stock market according to liquidity indices baxsed on the relative matching coefficient has been done. For this purpose, 42 companies listed on the stock exchange during the years 1397-1398 baxsed on the index selected liquidity options include turnover, proportional bid-ask spread, modified version of the amihud illiquidity measure, financial leverage, company performance, coefficient of elasticity of tradin, effective spread, turnover-adjusted number of zero daily volumes, number of trades, transaction volum percentage of transaction days,the size of the company was clustered baxsed on the Complete-lixnkage Clustering algorithm and the similarity coefficient of relative matching. The results of this study show that the indicators of number of trades and volume of trades can be used to predict liquidity.