This paper proposes a system that presents outlines of discussions in other groups in the same classroom as co-occurrence networks that reflect the activeness of discussions. The proposed system focuses on discussions conducted by two people during a lecture. The activeness of the discussions is analyzed using nonlinguistic acoustic information, which is calculated based on participants utterances collected by wearable devices. The co-occurrence networks are drawn with emphasizing the active parts of discussions by each group. We conducted an experiment to verify the effect to the participants by applying the proposed system to a lecture, and the effect of presenting the co-occurrence networks was examined in a subjective evaluation. The participants conducted another discussion after observing the co-occurrence networks of the previous discussion, and the second discussion was scored by the lecturer. As a result, the co-occurrence networks reflecting discussion activeness were not evaluated higher than those that did not reflect discussion activeness. This suggests that the variety of topics in the co-occurrence networks may stimulate discussion participants more effectively.