The parameter adjusting section 13 replaces distributed representations of the words included in the word space of the target language and common to the words included in the word space of the reference language with distributed representations of the words included in the word space of the reference language and corresponding to the common words included in the word space of the target language, and adjusts a parameter for a technique for producing a distributed representation of a word. For example, the parameter adjusting section 13 receives the target language learning corpus 22, selects the words included in the target language learning corpus 22 and common to the words included in the word space of the reference language, and replaces the distributed representations of the selected words with the distributed representation of the words included in the word space of the reference language. For example, the parameter adjusting section 13 replaces the distributed representations in the hidden layer of the Skip-gram model with the distributed representations of the words included in the word space of the reference language. Then, the parameter adjusting section 13 adjusts the weight W′_N×V that is the parameter between the hidden layer and the output layer. The parameter adjusting section 13 sequentially selects all the common words and executes a process of adjusting the parameter on the selected words.