This paper concerns estimating the user reference clusters for a database which can be used in partitioning a relational database horizontally during a distributed database design. using the knowledge about the data the user queries are converted to equivalent queries by a proposed inference procedure the user reference clusters estimated from these revised queries are more precise than those which can be estimated from the original user queries an extension of the language of first-order calculus is developed for the presentation of the user queries and the knowledge-base the types of knowledge to be stored are discussed an example illustrates the way inference is carried out, and the soundness of the inference is also discussed.