Technische Universität Kaiserslautern (TUK)

TUK is a public research university in Kaiserslautern, Germany. I am currently pursuing my master's degree in intelligent systems from this university.

American International University-Bangladesh (AIUB)

AIUB is considered as one of the top private university in Bangladesh. I completed my Bachelor of Science degree from Faculty of Science & Information Technology in CSE and placed at top 7%. At the time of my undergraduate studies, I am very much fortunate to have 3 years of experience as a “Student Assistant” in VUES (Virtual University & Experts System) department at AIUB.

Chittagong Public School & College

Chittagong Public School & College is also known as Chittagong Cantonment Public College (CCPC). It is a higher secondary educational institution in Chittagong, Bangladesh and administered by the Bangladesh Army. I completed my higher secondary education from Science group there.

Govt. Muslim High School

It is a higher secondary educational institution in Chittagong, Bangladesh and administered by the Bangladesh government. I completed my secondary education from Science group there.

Kolpojontro Foundation

Kolpojontro foundation is a non-profitable social research & welfare organization.

ICDL 2021 - International Conference on Development and Learning

Abstract: Humans are aware of the context and the environ-ment while communicating with an interaction partner. Althoughequipping a robot with such cognitive skills is still a far cry,integrating contextual information into a robot is a step towardsmaking Human-Robot Interaction (HRI) more meaningful, re-alistic or “human-like”. This paper focuses on establishing HRIbased on semantic and contextual information in the conver-sation. The humanoid robot called ROBIN recognizes the userutterances using a speech recognizer called Julius. The approachproposed extracts named entities, contextual information and thesentiment of each user utterance, with robot’s responses triggeredby a chatbot while gestures and facial expressions invoked bythe sentiment labels of the user speech. The human-centeredevaluation results based on the experimental scenarios show thatthe approach ensures more fluid and natural interaction.

ICAR 2021 - 20th IEEE International Conference on Advanced Robotics

Abstract: This paper proposes a computational approach to spot conversational topics for socially interactive robots. A speech recognition system called Julius has been used to transcribe the audible speech of an interlocutor to text. A rule-based topic-aware chatbot system coupled with a real-time web-based information extraction system have been employed to generate appropriate speech outcome for a humanoid robot named ROBIN. The conversational topic rules have been implemented on top of the existing chatbot system. Wikipedia contents have been extracted, allowing the robot to fetch factual information on-the-fly. Bodily movements of the robot have been generated based on the emotional state of the speech delivered by a human interaction partner. Additionally, the robot's vision system has been utilized so as to trigger event-specific behaviour out of the robot. The robot has been used in various interaction scenarios to establish topically-relevant affective interaction with humans. The experimental results show promising improvement over the earlier interaction system as far as the incorporation of a topic-aware conversation system is concerned..

RAAD 2021 - Conference on Robotics in Alpe-Adria-Danube Region

Abstract: The assessment of emotion is a complex process that encompasses information from various channels or sources, e.g., verbal, para-linguistic, non-verbal or even textual cues. Understanding the emotion of an interaction partner and exhibiting proper bodily reactions play a crucial role in establishing an engaging interaction. The evaluation of emotion, in general, is significant as far as emotionally intelligent human-robot interaction is concerned. In this paper, a text-based interaction system has been introduced to establish affective communication. A chatbot system has been implemented on a humanoid robot named ROBIN. The sentiment of the user input texts has been analyzed with MITIE machine learning toolkit. The robot considers emotional states of the users' text utterances and reacts with appropriate speech, gesture, and facial expressions. This triggers real-time affective behavior from the robot's perspective. The proposed system has been tested on several interaction scenarios to validate the approach from a human-centered perspective.

Technische Universität Kaiserslautern (TUK)

Abstract: Social assistive robots must have human-like behavior and capabilities to effectively communicate with the human that feel natural and engaging for interaction partner. With technological development in recent years, researches are trying to make more progress in Human-Robot Interaction (HRI) so that it can be more convincing and credible. The important characteristics of such meaningful conversation includes generating the robot's response that is topically relevant to the conversation and inclusion of affective expressions. To generate topically relevant response, robot needs to understand the inherent meaning of the conversation, and to be emotionally intelligent, it is important to detect the emotions of robot's interaction partner.
Guided by psychological concepts and ideas, this thesis contributes to the development of a new approach to generate robot's response in a human-robot interaction. A upper-torso humanoid robot, ROBIN recognizes the user utterances using a speech recognizer named ``Julius''. The proposed system recognizes the topic of the conversation from the human's utterance and generate relevant replies on-the-fly based on conversational rules. The topic detection module uses the information extraction module to detect the object and action from the human utterance to generate topically relevant reply. It also extracts the named entities, binary relationship between entities and the sentiment of each user utterance, with robot's response triggered by a chatbot while gestures and facial expressions invoked by the sentiment labels of the user speech.
The human-centered evaluation results based on the experimental scenarios shows that the approach ensures more fluid and natural interaction. It is concluded that the proposed system fulfills its goals, and suggestions for future work are presented.

American International University-Bangladesh (AIUB)

Abstract: Over the years stock market remains a gold mine for the investors as it is a high return on investment. The stock market offers a great return on investment but also promises a high amount risk due to its eccentric nature. Therefore much more research needs to be done for making a smart move in the stock market. Markov models and Conditional Probability over the time helps investors in taking decisions and forecasting the future of the investment. Investing in the stock market can equally be predicted under this two models. News can be very informative if they are analyzed in a proper way. Guessing a news actual sentiment can be very productive for an investor if he or she compiles to make most of their investment. We investigated multiple types of research on the stock market that uses different efficient mining techniques, natural language processing procedures, and AI techniques. In our work, we combined sentimental analysis and two different mathematical models to forecast the future trend based on the published authentic news. To conduct this research we chose Dhaka stock exchange and for the data resource we depended on DSE‟s official website.