This　paper　presents　adaptive　linear　market　value　functions　to　solve　the　problem　of　identification　of　customers　having　potential　market　value　in　targeted　marketing.　The　performance　of　these　methods　is　compared　with　some　standard　data　mining　methods　such　as　simple　Naive　Bayes.　Experiments　on　real　world　data　show　that　the　proposed　methods　are　efficient　and　effective.