To bridge the information gap in low-income countries, a critical issue is of being able to provide easily accessible information to low-literacy users. In particular, facilitating access to Science, Technology, Engineering and Mathematics (STEM) related information is crucial to bolster curiosity and accompany the democratization of engineering education. In order to tackle this challenge, we introduce MyOracle, a question-answering application focusing on STEM related questions and which is tailored to low-literacy users. MyOracle achieves this using three steps: extracting answers from the web, curating these answers and then summarizing and translating the resulting text. Finally we study and compare the disparities in content availability between nine languages with varying number of speakers and discuss the trade-offs between a translation-based approach and a native approach for lower-resource languages.

The provision of agricultural market information has demonstrated potential to improve the incomes of smallholder farmers who comprise over 80% of India’s agricultural population. The scarcity of accurate price information, however, leads farmers to accept suboptimal economic outcomes. By eliminating information asymmetries, the provision of actionable information theoretically reduces costs of price discovery and increases farmers’ bargaining power. The resulting price convergence increases market efficiency and reduces volatility. Evaluations of market information services, however, show mixed results. In addition to market barriers, technical and language illiteracy undermine the success of market information services in low-income countries. In this paper, we present Mobile Mandi – a mobile application that delivers daily retail prices of 22 essential commodities at 157 markets sourced from the Government of India’s Department of Consumer Affairs. We present next-day price forecasts by training long short-term memory (LSTM) models on a univariate time series dataset of daily crop prices over ten years and benchmark performance against a baseline persistence model. We design the user interface for low-literate users and provide a conversational agent to facilitate information dissemination in English, Hindi, Punjabi and its transliterations. Noting the prevalent use of the WhatsApp messaging service in India, we integrate the service for seamless interoperability.

The United Nation's 4th Sustainable Development Goal aims to provide quality education to all. Unfortunately, currently it is too often the case that learners from Low-Income Countries (LICs) cannot access relevant educational resources. Low bandwidth, low literacy and different cultural backgrounds often cause barriers of entry to access to education for learners in LICs.Conversely, learning environments have changed dramatically in the last decade. With the rise of eLearning platforms such as Coursera and edX, a learner can now stream high quality Open Educational Resources (OERs) right from their own home, as long as they have a stable Internet connection. With over 100 million participants worldwide, eLearning platforms provide alternatives to conventional classroom education, democratising education for everyone. But do these new learning environments really improve access to education in LICs?Research shows that unforeseen differences in Internet access, language, literacy, and cultural backgrounds between learners from High-Income and Low-Income Countries (HICs/LICs) mean that eLearning is often inaccessible for people in LICs. In this paper, we describe Word2Mouth, an eLearning platform that has been designed specifically to overcome the barriers of entry to eLearning for LICs, educating learners in LICs in a way that current eLearning platforms fail to do.

Drop out rates in Massive Open Online Courses (MOOCs) are very high. Although there are many external factors, such as users not having enough time or not intending to complete the course, there are some aspects that instructors can control to optimally engage their students. To do this, they need to know how students engage throughout their video lecture. This paper explores the use of webcams to record students’ facial expressions whilst they watched educational video material to analyse their Learner Engagement levels. Convolutional neural networks (CNNs) were trained to detect facial action units, which were mapped onto two psychological measurements, valence (emotional state) and arousal (attentiveness), using support vector regressions. These valence and arousal values were combined in a novel manner resulting in Learner Engagement levels. Moreover, a new approach was used to combine CNNs with geometric feature-based techniques to improve the performance of the models. Two experiments were conducted and found that 9 out of 10 CNN models achieved 95% accuracy on average across the majority of the subjects, whilst the Learner Engagement detector was able to identify facial expressions that translated to Learner Engagement levels successfully. These results suggest that there is promise in this approach, in that feedback on students’ Learner Engagement can be provided back to the instructor. Additional research should be undertaken to further prove these results and overcome some limitations that were faced.

Computing has recently been introduced as a core subject in British schools, meaning that children need to learn computer programming. Teachers have to be prepared to be able to deliver the new curriculum, but many of them do not feel confident teaching it as they have no formal background in Computer Science. Also, when learning to programme, children need the correct environment and support to succeed. This paper presents TuringLab, an environment to assist teachers in delivering the practical elements of the computing curriculum, while also proving to be engaging and challenging for the children. Teachers can create programming challenges for their pupils and see how they are progressing (or struggling) during completion of the challenges. Students can undertake challenges in an engaging environment which displays a graphical output of their code and assists in understanding errors they may encounter. TuringLab has been used to teach children how to programme at a number of volunteer-led coding clubs. Children engaged well with TuringLab, and the volunteers, who acted as teachers in these sessions, found TuringLab an extremely valuable educational tool.

A major driver of healthcare cost is the inefficiencies associated keeping less-critical patients overnight in the hospital when these patients could be treated as outpatients and received comparable quality of care. To this end, we propose the concept of tether-free patient to support patient mobility and ensure privacy-preserving healthcare. The paper addresses issues related to the design of a ubiquitous architectures to support tetherless care and proposes a robust, scalable and secure framework to location management of tether-free patients in mobile, wireless environments. The main building blocks and functionalities of the proposed architecture are described, and the algorithms used for secure registration and discovery, as the patient roams around, are presented. The complexity analysis of the proposed schemes is discussed.

As the complexity and size of home networks increases, so does the need for better management support for nonexpert home users. In this paper we describe an extensible policy-based configuration service that allows the home networking infrastructure to be viewed, verified and reconfigured at a higher-level of abstraction. The aim is to ensure that configuration changes in network infrastructure do not result in erroneous setups and to allow intuitive user-centric tools to operate at a higher-level of abstraction. Our implementation compiles configuration tasks to host/device level commands and supports deployment to multiple heterogeneous devices. We demonstrate and evaluate our approach using access control and bandwidth sharing as two examples of home management tasks.

Many missions are deemed dangerous or impractical to perform by humans, but can use collaborating, self-managing unmanned autonomous vehicles (UAVs) which adapt their behaviour to current context, recover from component failure or optimise performance. This paper describes a policy-based distributed self-management framework for both individual and teams of UAVs. We use three levels of specifications - policy, mission class and mission instance to enable reuse of both policies and mission classes. The architecture has been tested on devices ranging from small laptops to body area networks. Initial evaluation shows the distributed architecture is scalable and outperforms a centralised mission management scheme.

Sensor networks will shortly consist of globally deployed sensors providing real-time geo-centric information to users. Particularly, users with mobile devices will issue ad-hoc queries usually from within, or nearby, the queried area. In this paper, we propose an in- network data-centric storage (DCS) scheme, namely the spatio-temporal data-centric storage (STDCS) scheme, to efficiently answer these mobile user queries. STDCS is designed to maintain load-balancing among sensors to cope with query hotspots in the network. It is different from previous proposals in two aspects. First, it is based on the novel idea of using a temporally evolving spatial indexing scheme to balance querying load among sensors. Furthermore, STDCS uses dynamic mechanisms for query hotspot detection and decomposition. We conducted extensive simulations that showed our scheme's superiority to both local storage and spatial indexing (the only known geo-centric storage schemes) whenever experiencing query hotspots of different sizes.

The expected architecture of next-generation sensornets raises the need for load-balanced point-to-point routing protocols to cope with different sources of traffic skewness. Query hotspots are one of these skewness sources that highly impact the sensornet performance. In this paper, we present a set of content-based load-balancing primitives to be used on top of any point- to-point routing protocol in order to detect and decompose query hotspots. Our schemes are based on local hotspot detection by the sensors targeted by queries. Hotspots are then decomposed by avoiding duplication when forwarding results to the query issuers. Our simulation results show the high benefit, in terms of network lifetime and throughput, of using our schemes to load-balance query hotspots of different sizes.