It has been suggested that children learn the meanings of words by observing the regularities across different situations in which a word is used. However, experimental studies show that children are also sensitive to the syntactic properties of words and their context at a young age, and can use this information to find the correct referent for novel words. We present a unified computational model of word learning which integrates cross-situational evidence with the accumulated semantic properties of the lexical categories of words. Our experimental results show that using lexical categories can improve performance in learning, particularly for novel or low-frequency words in ambiguous situations.