Surprisingly, there was only a few research on Indonesian Natural Language Processing. There were two international projects on machine translation systems involving Bahasa Indonesia1 (MMTS and UNL projects). Although both projects did Indonesian morphological analysis also, only a part of prefix and suffix was taken into consideration since the number of affixes were too many to handle [1][2]. Chasen, a popular Japanese morphological analysis program uses connectivity costs to analysis japanese words and to solve ambiguous words[3][4].