Then, the learning termination determining section 15 determines whether or not the requirement for the termination of the learning is satisfied (in step S60). For example, the learning termination determining section 15 determines whether or not the difference between the weight W′_N×V that is the parameter before the update of the weight W′_N×V and the weight W′_N×V that is the parameter after the update of the weight W′_N×V is smaller than the threshold. The weight W′_N×V before the update is the parameter adjusted by the parameter adjusting section 13. The weight W′_N×V after the update is the parameter adjusted by the adjusted parameter distributed representation learning section 14. When the learning termination determining section 15 determines that the requirement for the termination of the learning is not satisfied (No in step S60), the learning termination determining section 15 causes the learning process to proceed to step S40 in order to further adjust the weight W′_N×V after the update.

When the learning termination determining section 15 determines that the requirement for the termination of the learning is satisfied (Yes in step S60), the learning termination determining section 15 terminates the learning process.