Rough　set　theory　is　an　efficient　information　processing　tool　used　in　the　discovery　of　data　dependencies.　It　evaluates　the　importance　of　attributes,　discovers　the　patterns　of　data,　reduces　all　redundant　objects　and　attributes,　and　seeks　the　minimum　subset　of　attributes.　This　paper　presents　a　method　for　attribute　reduction　on　combination　of　rough　set　and　neighborhood　systems.　Neighborhood　decision　system　is　investigated　by　considering　relation　between　two　ways　and　introducing　two　neighborhood　approximation　operators.　Illustrative　results　for　some　databases　in　UCI　repository　of　machine　learning　databases　provided　good　results.