Taranpreet Singh Saini

Bio: Taranpreet Singh Saini is an academic researcher from Massachusetts Institute of Technology. The author has contributed to research in topics: Affective computing & User requirements document. The author has an hindex of 2, co-authored 7 publications receiving 12 citations. Previous affiliations of Taranpreet Singh Saini include Maharashtra Institute of Technology.

Analysing human feelings by Affective Computing - survey

Abstract: Affective computing is one of the active research topic getting a lot of research attention recently. This increase in research interest is driven by many areas that are being worked on such as machine-based fact finding, smart over-seeing, perceptual connection, and so on. Identifying or deducing the feelings while they are on a particular task has a multidisciplinary domain involvement. This paper mainly emphasizes on basics of computing feelings while they are in session.

Inferring User Emotions from Keyboard and Mouse

Abstract: This chapter emphasizes on retrieving user emotions from keyboard and mouse using different parameters. These parameters can be user keyboard typing style, mouse movements, and some physiological sensors are used. This field of retrieving emotions from machines comes under the field of affective computing.

Inferring emotional state of a user by user profiling

Abstract: User profiles are important in many areas in which it is essential to obtain knowledge about users of software applications. Knowledge about a user includes his likes, dislikes, even his emotional state can be determined by user profiling. In this paper we examine what information constitutes a user profile; and how the profile information is used to get the emotional state of a user. We also study the main issues regarding user profiles from the perspectives of these research fields.

Inferring user emotions using physiological signals from mouse and keyboard

Abstract: In this paper, we have developed a mouse and keyboard which contains heart beat sensor, temperature sensor and force sensor. These sensors will generate the physiological signals. The signals from these devices will provide us the result for the current user. These results will be processed by the microcontroller and transmitted to the android device through Bluetooth module. The results will help us to determine the user's current emotional state and if found negative it can be altered.

Automobile driver profiling: Infering emotions through actions for safe driving

Abstract: Driving an automobile taxes the driver in almost all of his sensory systems and uses his deduction power to the fullest. The mental state of the driver is vital to safe and correct driving. This paper focuses on identifying the emotional state of the driver based on his interactions with the automobile while driving and inferring when the driver is in an unpleasant state of mind. The driver's actions are monitored and driver profiles are generated which lead to identifying his usual behavior. If the behavior of the driver, based on his actions and interactions with the automobile, implies he is in a negative state of mind, we propose to deviate his mind from negative to positive emotions. For this we have instrumented a system which is both single mode and multimodal to alert the driver with includes feedback in the form of auditory, haptic and visual.

AGaR: A VR Serious Game to Support the Recovery of Post-Stroke Patients

Abstract: Games can make training procedures more engaging for patients. Considering the complexity of the process for upper limb function rehabilitation, this paper presents the development and an initial evaluation of the AGaR – a serious game with virtual reality and natural interaction, both to aid patients to execute repetitive exercises and to aid physiotherapists to follow the rehabilitation process. Additionally, we obtain and analyze data about patients' engagement as a differential in relation to others games developed for similar goals. In this game, the patient has to associate two different images with complementary meanings, using a movement sensor to drag the image to the target. To evaluate the game, an initial experiment was conducted with patients. The results show that, within the rounds played by the participants of the experiment, the number of wrong associations made by them varies according to patient, with no standard found. The engagement tends to increase during use of the game, throughout the rounds.

Energy-Efficient Joint Resource Allocation Algorithms for MEC-Enabled Emotional Computing in Urban Communities

Abstract: This paper considers a mobile edge computing (MEC) system, where the MEC server first collects data from emotion sensors and then computes the emotion of each user. We give the formula of the emotional prediction accuracy. In order to improve the energy efficiency of the system, we propose resources allocation algorithms. We aim to minimize the total energy consumption of the MEC server and sensors by jointly optimizing the computing resources allocation and the data transmitting time. The formulated problem is a non-convex problem, which is very difficult to solve in general. However, we transform it into convex problems and apply convex optimization techniques to address it. The optimal solution is given in closed form. Simulation results show that the total energy consumption of our system can be effectively reduced by the proposed scheme compared with the benchmark.

Human Emotion Recognition using Physiological Signals: A Survey

Abstract: Human emotions are one of the ways to express our feelings. Affective Computing originates from the study of human emotions. Over the years, psychologists have developed various emotional models to explain the emotional or affective states of humans. Affective Computing uses various models of emotion and machine learning algorithms to classify emotions. Machine Learning enables computers to learn from the training datasets and classify new input, thus it can be effectively used to teach computers to understand human emotions. This paper focuses on a survey of human emotion recognition using physiological signals related to the human body like Electrocardiogram (ECG), Electroencephalogram (EEG), Electromyogram (EMG), Galvanic Skin Response (GSR), Respiration (RSP), Skin Temperature (SKT), etc. and also their advantages and disadvantages. It also describes challenges in physiological sensing for Affective Computing.