As the number of people receiving psycho therapeutic treatment increases, the au tomatic evaluation of counseling practice arises as an important challenge in the clinical domain. In this paper, we address the automatic evaluation of counseling performance by analyzing counselors' lan guage during their interaction with clients. In particular, we present a model towards the automation of Motivational Interview ing (MI) coding, which is the current gold standard to evaluate MI counseling. First, we build a dataset of hand labeled MI en counters; second, we use text-based meth ods to extract and analyze linguistic pat terns associated with counselor behaviors; and third, we develop an automatic sys tem to predict these behaviors. We intro duce a new set of features based on seman tic information and syntactic patterns, and show that they lead to accuracy figures of up to 90%, which represent a significant improvement with respect to features used in the past.
展开▼