Data Mining - Concepts and Techniques.

Data Mining Practical Machine Learning Tools and Techniques

As we are moving towards the Internet of Things (IoT), the number of sensors deployed around the world is growing at a rapid pace. Market research has shown a significant growth of sensor deployments over the past decade and has predicted a significant increment of the growth rate in the future. These sensors continuously generate enormous amounts of data. However, in order to add value to raw sensor data we need to understand it. Collection, modelling, reasoning, and distribution of context in relation to sensor data plays critical role in this challenge. Context-aware computing has proven to be successful in understanding sensor data. In this paper, we survey context awareness from an IoT perspective. We present the necessary background by introducing the IoT paradigm and context-aware fundamentals at the beginning. Then we provide an in-depth analysis of context life cycle. We evaluate a subset of projects (50) which represent the majority of research and commercial solutions proposed in the field of context-aware computing conducted over the last decade (2001-2011) based on our own taxonomy. Finally, based on our evaluation, we highlight the lessons to be learnt from the past and some possible directions for future research. The survey addresses a broad range of techniques, methods, models, functionalities, systems, applications, and middleware solutions related to context awareness and IoT. Our goal is not only to analyse, compare and consolidate past research work but also to appreciate their findings and discuss their applicability towards the IoT.

/pdf/context-aware-computing-for-the-internet-of-things-a-survey-3nibvvc921.pdf

Context Aware Computing for The Internet of Things: A Survey

Matrix Factorization Techniques for Recommender Systems

We quantitatively investigate how machine learning models leak information about the individual data records on which they were trained. We focus on the basic membership inference attack: given a data record and black-box access to a model, determine if the record was in the model's training dataset. To perform membership inference against a target model, we make adversarial use of machine learning and train our own inference model to recognize differences in the target model's predictions on the inputs that it trained on versus the inputs that it did not train on. We empirically evaluate our inference techniques on classification models trained by commercial "machine learning as a service" providers such as Google and Amazon. Using realistic datasets and classification tasks, including a hospital discharge dataset whose membership is sensitive from the privacy perspective, we show that these models can be vulnerable to membership inference attacks. We then investigate the factors that influence this leakage and evaluate mitigation strategies.

/pdf/membership-inference-attacks-against-machine-learning-models-2tzchejf91.pdf

Membership Inference Attacks Against Machine Learning Models

Enabling Smart Urban Services with GPS Trajectory Data

Context information has emerged as an important resource to enable autonomy and flexibility of ubiquitous applications. The widespread use of context information necessitates an efficient lookup service in a wide-area network over multiple smart spaces. In this paper, we propose a context lookup framework based on a semantic peer-to-peer network to support the building of context-aware applications in multiple smart spaces. Collaborative contextaware applications that utilize different context information in multiple smart spaces can be easily built by invoking a pull or push service provided by our framework. We describe the design of our system, demonstrate the development process of context-aware applications, and report the measurements obtained from our prototype.

/pdf/a-semantic-p2p-framework-for-building-context-aware-42y54pyz1x.pdf

A semantic P2P framework for building context-aware applications in multiple smart spaces

With the proliferation of increasingly powerful mobile devices and wireless networks, mobile crowdsourcing has emerged as a novel service paradigm. It enables crowd workers to take over outsourced location-dependent tasks, and has attracted much attention from both research communities and industries. In this paper, we consider a mobile crowdsourcing scenario, where a mobile crowdsourcing task is too complex (e.g., post-earthquake recovery, citywide package delivery) but can be divided into a number of easier subtasks, which have interdependency between them. Under this scenario, we investigate an important problem, namely task graph scheduling in mobile crowdsourcing (TGS-MC), which seeks to optimize a compact scheduling, such that the task completion time (i.e., makespan) and overall idle time are simultaneously minimized with the consideration of worker reliability. We analyze the complexity and NP-complete of the TGS-MC problem, and propose two heuristic approaches, including BFS-based dynamic priority scheduling BFSPriD algorithm, and an evolutionary multitasking-based EMTTSch algorithm, to solve our problem from local and global optimization perspective, respectively. We conduct extensive evaluation using two real-world data sets, and demonstrate superiority of our proposed approaches.

Compact Scheduling for Task Graph Oriented Mobile Crowdsourcing

With the growing number of the elderly, we see a greater demand for home care, and the vision of pervasive computing is also floating into the domain of the household that aims to build a smart home which can assist inhabitants (users) to live more conveniently and harmoniously. Such health-care pervasive applications in smart home should focus on the inhabitant’s goal or task in diverse situations, rather than the various complex devices and services. The core challenge for homecare design is to perceive the environment and assess occurring situations, thus allowing systems to behave intelligently according to the user’s intent. Due to the dynamic and heterogeneous nature of pervasive computing environment, it is difficult for an average user to obtain right information and service and in right place at right time. This paper proposes a context-driven task-oriented middleware (CDTOM) to meet the challenge. The most important component is its task model that provides an adequate high-level description of user-oriented tasks and their related contexts. Leveraging the model multiple entities can easily exchange, share and reuse their knowledge. Based on the hierarchy of task ontology, a novel task recognition approach using CBR (case-based reasoning) is presented and the performance of task recognition is evaluated by task number, context size and time costing. Moreover, a dynamic mechanism for mapping the recognized task and services is also discussed. Finally, we present the design and implementation of our task supporting system (TSS) to aid an inhabitant’s tasks in light of his lifestyle and environment conditions in pervasive homecare environment, and the results of the prototype system show that our middleware approach achieves good efficiency of context management and good accuracy of user’s activity inference, and can improve efficiently quality of user’s life.

CDTOM : a context-driven task-oriented middleware for pervasive homecare environment

Store site recommendation aims to predict the value of the store at candidate locations and then recommend the optimal location to the company for placing a new brick-and-mortar store. Most existing studies focus on learning machine learning or deep learning models based on large-scale training data of existing chain stores in the same city. However, the expansion of chain enterprises in new cities suffers from data scarcity issues, and these models do not work in the new city where no chain store has been placed (i.e., cold-start problem). In this article, we propose a unified approach for cold-start store site recommendation, Weighted Adversarial Network with Transferability weighting scheme (WANT), to transfer knowledge learned from a data-rich source city to a target city with no labeled data. In particular, to promote positive transfer, we develop a discriminator to diminish distribution discrepancy between source city and target city with different data distributions, which plays the minimax game with the feature extractor to learn transferable representations across cities by adversarial learning. In addition, to further reduce the risk of negative transfer, we design a transferability weighting scheme to quantify the transferability of examples in source city and reweight the contribution of relevant source examples to transfer useful knowledge. We validate WANT using a real-world dataset, and experimental results demonstrate the effectiveness of our proposed model over several state-of-the-art baseline models.

Daqing Zhang

Papers

Enabling Smart Urban Services with GPS Trajectory Data

A semantic P2P framework for building context-aware applications in multiple smart spaces

Compact Scheduling for Task Graph Oriented Mobile Crowdsourcing

CDTOM : a context-driven task-oriented middleware for pervasive homecare environment

Knowledge Transfer with Weighted Adversarial Network for Cold-Start Store Site Recommendation