Objectives and expected results

In humans the concepts of reasonable and safe actions can be taught directly via natural language. In this project we will develop an understanding of explicit information from dialogues in natural language to infer and learn safe and unsafe concepts. In a conversation, humans use changes in a dialogue to predict safety-critical situations and use them to react accordingly. We propose to use the same cues for safer human-robot interaction for early verbal detection of dangerous situations. To achieve this goal, we use different language features, such as sentiment and dialogue act. We developed the neural models that learn to predict the sentiment of next upcoming utterance and to recognize the dialogue act of that utterance. Currently, we aim to bind them in a way to achieve our primary goal.

Keywords: Natural Language Processing, Human-Robot Interaction, Sentiment Analysis, Dialogue Act Processing

Short Curriculum


  • Since March 2016: Research Associate (PhD Student: SECURE Project) at Knowledge Technology Research Group, Department of Computer Science, University of Hamburg, Germany.
    • PhD Title: Learning and generalizing safety concepts through language processing for human-robot verbal interaction
  • Qualification:
    • September 2015: Master in "Robotics and Applied Informatics", Ecole Centrale de Nantes, Nantes, France.
      Thesis Title: "Human-Humanoid Interaction by Verbal Dialogues"
    • June 2011: Bachelor of Enggineering in "Electronics and Tele-Communication", Aunradha Engg. College (Chikhli), Amravati University (MS), Amravati, India.

Projects Grants

Activities: Workshops, school and other participations


Discourse-Wizard Live-Demo: Discovering Deep Discourse Structure in your Conversation with RNNs.
Dialogue Act Recognition Demonstration with and without context model, shows the importance of context in a conversation. [initial release: 25 May, 2018]