Search

As the innovation of smart devices and internet-of-things (IoT), smart homes have become prevalent. People tend to transform residences into smart homes by customizing off-the-shelf smart home platforms, instead of creating IoT systems from scratch. Among the alternatives, Home Assistant (HA) is one of the most popular platforms. It allows end-users (i.e., home residents) to smartify homes by (S1) integrating selected devices into the system, and (S2) creating YAML files to control those devices. Unfortunately, due to the diversity of devices and complexity of automatic configurations, many users have difficulty correctly creating YAML files. Consequently, their smart homes may not work as expected, causing frustration and concern in users. This paper presents a novel study on issues of YAML-based automation configuration in smart homes (issues related to S2). We mined the online forum Home Assistant Community for discussion threads related to automation configuration. By manually inspecting 190 threads, we revealed 3 categories of concerns: implementation, optimization, and debugging. Under each category, we classified discussions based on the issue locations and technical concepts involved. Among debugging discussions, we further classified discussions based on users' resolution strategies; we also applied existing analysis tools to buggy YAML files, to assess the tool effectiveness. Our study reveals the common challenges faced by users and frequently applied resolution strategies. There are 129 (68%) examined issues concerning debugging, but existing tools can detect at most 14 issues and fix none. It implies that existing tools provide limited assistance in automation configuration. Our research sheds light on future directions in smart home development., Comment: 13 pages, 3 figures, 3 tables, 10 listings

The paper describes a dataset comprising indoor environmental factors such as temperature, humidity, air quality, and noise levels. The data was collected from 10 sensing devices installed in various locations within three single-family houses in Virginia, USA. The objective of the data collection was to study the indoor environmental conditions of the houses over time. The data were collected at a frequency of one record per minute for a year, combining over 2.5 million records. The paper provides actual floor plans with sensor placements to aid researchers and practitioners in creating reliable building performance models. The techniques used to collect and verify the data are also explained in the paper. The resulting dataset can be employed to enhance models for building energy consumption, occupant behavior, predictive maintenance, and other relevant purposes.

The user persona is a communication tool for designers to generate a mental model that describes the archetype of users. Developing building occupant personas is proven to be an effective method for human-centered smart building design, which considers occupant comfort, behavior, and energy consumption. Optimization of building energy consumption also requires a deep understanding of occupants' preferences and behaviors. The current approaches to developing building occupant personas face a major obstruction of manual data processing and analysis. In this study, we propose and evaluate a machine learning-based semi-automated approach to generate building occupant personas. We investigate the 2015 Residential Energy Consumption Dataset with five machine learning techniques - Linear Discriminant Analysis, K Nearest Neighbors, Decision Tree (Random Forest), Support Vector Machine, and AdaBoost classifier - for the prediction of 16 occupant characteristics, such as age, education, and, thermal comfort. The models achieve an average accuracy of 61% and accuracy over 90% for attributes including the number of occupants in the household, their age group, and preferred usage of heating or cooling equipment. The results of the study show the feasibility of using machine learning techniques for the development of building occupant persona to minimize human effort., Comment: 12 pages, 4 figures

Based on digital pathology slice scanning technology, artificial intelligence algorithms represented by deep learning have achieved remarkable results in the field of computational pathology. Compared to other medical images, pathology images are more difficult to annotate, and thus, there is an extreme lack of available datasets for conducting supervised learning to train robust deep learning models. In this paper, we propose a self-supervised learning (SSL) model, the global contrast-masked autoencoder (GCMAE), which can train the encoder to have the ability to represent local-global features of pathological images, also significantly improve the performance of transfer learning across data sets. In this study, the ability of the GCMAE to learn migratable representations was demonstrated through extensive experiments using a total of three different disease-specific hematoxylin and eosin (HE)-stained pathology datasets: Camelyon16, NCTCRC and BreakHis. In addition, this study designed an effective automated pathology diagnosis process based on the GCMAE for clinical applications. The source code of this paper is publicly available at https://github.com/StarUniversus/gcmae.

The deep neural network is a research hotspot for histopathological image analysis, which can improve the efficiency and accuracy of diagnosis for pathologists or be used for disease screening. The whole slide pathological image can reach one gigapixel and contains abundant tissue feature information, which needs to be divided into a lot of patches in the training and inference stages. This will lead to a long convergence time and large memory consumption. Furthermore, well-annotated data sets are also in short supply in the field of digital pathology. Inspired by the pathologist's clinical diagnosis process, we propose a weakly supervised deep reinforcement learning framework, which can greatly reduce the time required for network inference. We use neural network to construct the search model and decision model of reinforcement learning agent respectively. The search model predicts the next action through the image features of different magnifications in the current field of view, and the decision model is used to return the predicted probability of the current field of view image. In addition, an expert-guided model is constructed by multi-instance learning, which not only provides rewards for search model, but also guides decision model learning by the knowledge distillation method. Experimental results show that our proposed method can achieve fast inference and accurate prediction of whole slide images without any pixel-level annotations.

The vast number of facility management systems, home automation systems, and the ever-increasing number of Internet of Things (IoT) devices are in constant need of environmental monitoring. Indoor environment data can be utilized to improve indoor facilities and better occupants' working and living experience, however, such data are scarce because many existing facility monitoring technologies are expensive and proprietary for certain building systems, such as building automation systems, energy management systems, and maintenance systems. In this work, the authors designed and prototyped a cost-effective, distributed, scalable, and portable indoor environmental data collection system, Building Data Lite (BDL). BDL is based on Raspberry Pi computers and multiple changeable arrays of sensors, such as sensors of temperature, humidity, light, motion, sound, vibration, and multiple types of gases. The system includes a distributed sensing network and a centralized server. The server provides a web-based graphical user interface that enables users to access the collected data over the Internet. To evaluate the BDL system's functionality, cost-effectiveness, scalability, and portability, the research team conducted a case study in an affordable housing community where the system prototype is deployed to 12 households. The case study results indicate that the system is functioning as designed, costs about \$3500 to sense 48 building zones (about \$73 per zone) and provides 12 types of indoor environment data, is easy to scale up, and is fully portable., Comment: 32 pages, 15 figures

Traditionally, caregivers, whether formal or informal, have taken the responsibility of providing assistance and care to patients with cognitive decline. Usually, both the caregiver and the patient are subjected to financial and emotional burdens, which impact the patient's life quality. To overcome this issue, Ambient Assistive Living (AAL) technologies have been adopted to replace the caregivers and complement patients' lack of intelligence. Technologies such as Internet of Things (IoT) and Artificial Intelligence (AI) have enabled intelligent ubiquitous learning for smart buildings to monitor the cognitively declined occupants and provide in-home assistive services and solutions. This paper aims to summarize and evaluate the intelligence complements provided by smart buildings that can increase the cognitively declined occupants' quality of life and autonomy. Through a systematic literature review, the authors find that most of the existing contributions are towards identifying the occupants' behavior, and thus, to determine corresponding assistive services and solutions. Five key research gaps are identified, including the lack of adequate adoption of technological interventions to fully support the occupants' autonomy and independence. The authors also propose a conceptual framework to highlight the research gaps in smart building applications for cognitively declined occupants and to map the future research directions., Comment: 31 pages, 3 figures, 1 table. Submitted to the Journal of Building and Environment

In this paper, we study the Riley polynomial of double twist knots with higher genus. Using the root of the Riley polynomial, we compute the range of rational slope $r$ such that $r$-filling of the knot complement has left-orderable fundamental group. Further more, we make a conjecture about left-orderable surgery slopes of two-bridge knots., Comment: 24 pages, 7 figures

In Dunfield's catalog of the hyperbolic manifolds in the SnapPy census which are complements of L-space knots in $S^3$, we determine that $22$ have tunnel number $2$ while the remaining all have tunnel number $1$. Notably, these $22$ manifolds contain $9$ asymmetric L-space knot complements. Furthermore, using SnapPy and KLO we find presentations of these $22$ knots as closures of positive braids that realize the Morton-Franks-Williams bound on braid index. The smallest of these has genus $12$ and braid index $4$., Comment: This is a significant expansion/rewrite of the original version. The title and abstract have both changed to reflect this. Relevant code is included in the body of the article itself. Originally, this was a note that collected the results of an "office hour" session on the use of SnapPy and KLO held at the ICERM workshop Perspectives on Dehn Surgery, July 15--19, 2019

In this paper, we develop a method for constructing left-orders on the fundamental groups of rational homology 3-spheres. We begin by constructing the holonomy extension locus of a rational homology solid torus $M$, which encodes the information about peripherally hyperbolic $\widetilde{\text{PSL}_2\mathbb{R}}$ representations of $\pi_1(M)$. Plots of the holonomy extension loci of many rational homology solid tori are shown, and the relation to left-orderability is hinted. Using holonomy extension loci, we study rational homology 3-spheres coming from Dehn filling on rational homology solid tori and construct intervals of Dehn fillings with left-orderable fundamental group., Comment: 32 pages, 20 figures. arXiv admin note: text overlap with arXiv:1602.03793 by other authors