Joy Bose

TV remote control via wearable smart watch device

A number of wearable devices are becoming popular nowadays, giving rise to a number of possibilities for increased user convenience when doing everyday tasks. In this paper we propose a wearable smart watch with user customized gestures used to control a web browser on a smart TV running the Tizen Operating System. A mobile computing device, such as a smartphone, is used to translate the smart watch commands into a format that the TV can understand, as well as perform any other processing tasks. Since the smart watch is generally used by only one user, our system can incorporate user specific gestures to control the browser on the TV. Alternatively, the mobile phone can itself be held in the users hand instead of the smart watch, and used for making gestures. Such a system, using a smart watch or similar device for capturing the users gestures, is more user friendly than current systems on smart TVs that use a camera to track the hand gesture movements. We give the details of the implementation of the system and some tests to measure the accuracy of the detection of some hand gestures.

Attention Sensitive Web Browsing

The attention level is an important indicator of the users level of interest while viewing any content. The web browser is one of the most popular means to access information, and the usage of browsers in mobile devices is increasing. In this paper we analyze the use of attention as an input for web browsers. Attention can be measured easily in real time using cheap commercially available wearable EEG sensors, such as NeuroSkys MindWave. We use the measured level of attention in the following ways: as an input mechanism for navigating through the controls on the web browser such as buttons, menus and hyperlinks, to correlate the attention with the section of the webpage being browsed and make the web browser responsive to the users attention level in real time, and as an input that is fed back to the web server enabling the web content developer to make attention sensitive websites. For each of these, we provide the implementation details and some results obtained. We also provide some pointers how the input attention level event obtained from the EEG sensors can be standardized in the W3C specification.

Contextual adaptive user interface for Android devices

In this paper we propose a framework to adapt the user interface (UI) of mobile computing devices like smartphones or tablets, based on the context or scenario in which user is present, and incorporating learning from past user actions. This will allow the user to perform actions in minimal steps and also reduce the clutter. The user interface in question can include application icons, menus, buttons window positioning or layout, color scheme and so on. The framework profiles the user device usage pattern and uses machine learning algorithms to predict the best possible screen configuration with respect to the user context. The prediction will improve with time and will provide best user experience possible to the user. To predict the utility of our model, we measure average response times for a number of users to access certain applications randomly on a smartphone, and on that basis predict time saved by adapting the UI in this way.

Implementation and analysis of pluggable Android applications

In this paper we present an implementation for a typical Android application to support Plug-ins or add-on application to enhance functionality at run time. Such a framework allows an application to offer new functionality without updating the base application every time in the application store. Our implementation uses advanced bound services and AIDL concepts to provide communication between the application and its plugins. It implements an installed plugin discovery mechanism and also provides deployment from Plug-In SDK perspectives. The plug-ins developed with this framework implementation use the typical Android method for application development, which is the Android SDK. Therefore, no extra skill set is required for first or third party developers to write such plug-ins. We present how the plug-ins are discovered and launched from the parent application, and also how the parent application can expose APIs which are utilized by the plug-ins. We consider typical native Android applications as well as web applications as plugins. We identify key problems in implementing web applications as plugins and their solutions. We also analyze how to open the plug-in development kit to third party developers i.e. deployment scenarios of the plugin SDK. Finally, we determine the time taken to execute an API which transfers various sizes of data over bound service inter process communication.

Framework to Improve the Web Application Launch Time

Having too many applications running simultaneously on a smartphone consumes memory and slows down the performance of the device. Hence we need web applications which are lightweight and consume less memory. Web Applications use the browser engine and take a lot of time to launch compared to a native application, especially upon device boot up or if the browser is not already running in the background. In this paper, we propose an intelligent framework to launch web applications as fast as native applications. The framework considers the users usage of web applications and pre-launches the preferred web applications, thus enhancing the launch time performance. We provide the architecture and implementation details of the framework. We then perform experiments on various web applications to measure the effectiveness of the framework for fast launch of the applications after the device boots.

Seamless Push service with flow control for embedded devices

Push Notification Service is an essential service in smart phone platforms to give real time updates to users from Content providers. W3C Push API has defined a seamless way to enable real time updates to web applications using the web Push API. Content providers have to closely monitor which platform they are using and which Push provider they are using. An architecture that is agnostic to platforms and Push providers can open many features and services to users. In this paper, we present a platform independent architecture and methods to achieve Push message flow control suitable for embedded devices. Our architecture also gives a seamless flow continuity of Push messages for users. We present results of an experiment to measure the effectiveness of the system to perform flow control of the Push messages to ensure click rate of messages are increased.

EEG based detection of area of interest in a web page

We focus on the problem of detection of the users area of interest within a single web page, or the web page of interest within different web pages. Current methods either use some kind of manual ranking, or apply parameters such as the time the user spends on a specific area of the page to determine the area of interest. We postulate that the attention level of the user while browsing is a more reliable indication of the users level of interest. We use EEG inputs from a NeuroSky MindWave headset to capture the users attention level in real time. A background script in a web browser in a mobile device captures the part of the webpage currently being browsed by noting the percentage of the page that the user has scrolled to. The attention level and the percentage of the page scrolled are mapped using the timestamp as the key. Our solution is integrated with the mobile web browser architecture. Using our method, we determine and map the average attention level within the same page, and across different pages, for a range of websites and users. This can be useful in a number of applications including: providing inputs of user behavior to web developers for better web design, ranking different websites or videos as per user interest, inserting ads in the regions of a web page where the user is more likely to pay attention to.

A kiosk based model for employment generation in rural areas

Unemployment, especially among the youth, is a pressing issue in many rural areas. This paper proposes a model of local employment generation in rural or underdeveloped areas, which often do not have an adequate infrastructure (including wireless connectivity and uninterrupted electricity supply) to enable them to take advantage of technology. We assume that most people in such areas already possess and can make use of a basic mobile phone with facilities to call and text. Our model uses a kiosk based architecture (similar to ITCs e-Chaupal), with a subscription based model of services to standalone clients. The kiosks are equipped with net connectivity and enable the services to the mobile phones of the clients, with each kiosk serving up to 500 rural users. Users looking for work update their employable skills at the system, while users needing jobs done post the details. The system connects the skills with the places where needed, and includes the facilities to rank employment seeking users with their suitability for the job in real time. Such a model is feasible to run with optimal costs, and can play a good role in stimulating local employment and ultimately the local economy.

Secure web Push system

In existing push systems, there is no provision for additional security and the push messages from any content provider can be simply viewed by the user. This could cause problems in certain cases where additional security, privacy and authentication might be desirable before the user is allowed to view the push message, for example a message from the users bank giving the current bank balance. In this paper we present a Smart Push system that ensures user security and privacy. Our push architecture incorporates a Gateway Client and Gateway Server component, enabling additional security measures and encryption of sent push messages at the gateway server before the messages reach the client device. In our system, a unique subscription Id is generated during the registration process, which hides the actual registration Id of the client device and thus ensures privacy. An additional spam filter at the gateway server further protects the client device from spam push messages. We present the architecture of the system and results of some tests performed to measure the effectiveness of the security aspects of the push architecture.

Bias Based Navigation for News Articles and Media

In existing news related services, readers cannot decide if there is another side of a news story, unless they actually come across an article representing a different perspective. However, it is possible to determine the bias in a given article using NLP related tools. In this paper we determine the bias in media content such as news articles, and use this determined bias in two ways. First, we generate the topic/bias index for one or more news articles, positioning each article within the index for a given topic or attribute. We then provide a user interface to display how much bias is present in the currently read article, along with a slider to enable the reader to change the bias. Upon the user changing the bias value, the system loads a different news article on the same topic with a different bias. The system can be extended for a variety of media such as songs, provided the lyrics are known. We test our system on a few news articles on different topics, reconfirming the detected bias manually.