logo
Loading


New technology that makes life much easier for all of us

Improved sequence labeling enables smart homes and makes life much easier
For more information and details on this technology, email research@iitbmonash.org

New technology is being developed everyday to help make life easier, such as smart homes that turn appliances on and off at our direction, or speech recognition devices. These technologies are generally based on sequence labeling, which classifies how tasks occur in a consequential sequence.

Research being undertaken by Naveen Nair at the IITB-Monash Research Academy is looking at how to take sequence labeling to the next level, improving our understanding of how people perform activities and the relationship between each of their actions, so that the technology can become more efficient. For example, how can tasks be labeled in a smart home for the elderly so that it assists them by inferringtheir day to day actions.

Traditionally sequence labelling considers each action as part of a single variable and does not consider the overall context of the activity or the relationships between different actions that have been observed. Nair’s research focuses on finding the relationship between individual actions that have been observed. The learnings from these observations are then used to develop better inference models.

For example, each evening a person may open their bedroom door and then turn off the bedside heater. Where traditional sequence labeling would view these as two separate events, this research captures the relationship between the two actions, thereby resulting in a more meaningful system.

Nair’s research is focused on observing the activity in its entirety with the intention of being able to develop an optimum inference model for the tasks. For example, a person may open their bedroom door for several reasons. However if we have observed them open (and close) their bedroom door and turn off the bedside heater, then it is more likely that they are going to sleep.

The system developed by Nair automatically learns such relationships in an efficient and optimal way.This information could be used to optimise the way devices are utilised in a smart home. Whereby, when a person opens (and closes) their bedroom door and turns off the bedside heater, the system infers that the person is going to sleep and automatically switches off the television in the living room.

The research being undertaken by Naveen Nair is under the guidance of Professors Ganesh Ramakrishnan (from the Indian Institute of Technology, Bombay, India) and Shonali Krishnaswamy (from Monash University, Australia). IITB-Monash Research Academy is a Joint Venture between the IITB and Monash. Opened in 2008, the IITB-Monash Research Academy is a graduate research school located in Mumbai that aims at enhancing research collaborations between Australia and India. Students study for a dual PhD from both institutions, and spend time during their research in both India and Australia.

The research will have a variety of practical applications according to Naveen Nair “As my research is based on learning optimum features for structured output, it will improve accuracy and can be used to perform efficient activity recognition. In the observations I have made with elderly people living alone, the inference model developed will be able to estimate the health of a person (as activities of daily living are indicators of health condition). It may also aid systems that optimise device utilization. Other potential applications for this technology may be as broad as in named entity recognition whereby more accurate identification of words within text are then used to extract journalistic articles.”

With its wide range of practical applications, such as labeling activities in a smart home so as to optimise the utlisation of each device and minimise power consumption, this research is very exciting and has the potential to stimulate new research with further application.

Research Scholar: Naveen Nair, IITB-Monash Research Academy

Project Title: Feature Induction for Sequence Labeling

Supervisors: Professor Ganesh Ramakrishnan (IIT Bombay), Professor Shonali Krishnaswamy (Monash University)

Contact Details: naveennairs@gmail.com

Contact research@iitbmonash.org for more information on this, and other projects

The above story was written by Ms Rakhee Ghelani based on inputs from the research student and IITB-Monash Research Academy. Copyright IITB-Monash Research Academy.


^