With a huge amount of data generated every second, it has become important to remove data anomalies. Outliers are the extreme value that deviates from other observations in data. We propose a novel outlier detection method; FCBODM (Fuzzy Constraint based Outlier Detection Method) that takes into account of fuzzy constraint and background knowledge to discover the outliers in a dataset. Our key idea is to use fuzzy constraint technology wherein we used nearness measure theory in fuzzy mathematics for finding similarities between data objects and background information. It helps in finding more meaningful outliers. Our novel approach can be integrated with traditional outlier detection methods to improve the outlier ranking. In order to validate and demonstrate the effectiveness and scalability of our method we experimented it on real and semantically meaningful datasets. © 2019, Springer Nature Switzerland AG.