More Like this

1. Introducing Point-of-Interest as an alternative to Area-of-Interest for fixation duration analysis

   https://doi.org/10.1371/journal.pone.0250170  

   Rim, Nak Won; Choe, Kyoung Whan; Scrivner, Coltan; Berman, Marc G. (May 2021, PLOS ONE)

   Kasneci, Enkelejda (Ed.)

   Many eye-tracking data analyses rely on the Area-of-Interest (AOI) methodology, which utilizes AOIs to analyze metrics such as fixations. However, AOI-based methods have some inherent limitations including variability and subjectivity in shape, size, and location of AOIs. In this article, we propose an alternative approach to the traditional AOI dwell time analysis: Weighted Sum Durations (WSD). This approach decreases the subjectivity of AOI definitions by using Points-of-Interest (POI) while maintaining interpretability. In WSD, the durations of fixations toward each POI is weighted by the distance from the POI and summed together to generate a metric comparable to AOI dwell time. To validate WSD, we reanalyzed data from a previously published eye-tracking study (n = 90). The re-analysis replicated the original findings that people gaze less towards faces and more toward points of contact when viewing violent social interactions. 

   more » « less
2. GRID: A Python Package for Field Plot Phenotyping Using Aerial Images

   https://doi.org/10.3390/rs12111697  

   Chen, Chunpeng James; Zhang, Zhiwu (June 2020, Remote Sensing)

   Aerial imagery has the potential to advance high-throughput phenotyping for agricultural field experiments. This potential is currently limited by the difficulties of identifying pixels of interest (POI) and performing plot segmentation due to the required intensive manual operations. We developed a Python package, GRID (GReenfield Image Decoder), to overcome this limitation. With pixel-wise K-means cluster analysis, users can specify the number of clusters and choose the clusters representing POI. The plot grid patterns are automatically recognized by the POI distribution. The local optima of POI are initialized as the plot centers, which can also be manually modified for deletion, addition, or relocation. The segmentation of POI around the plot centers is initialized by automated, intelligent agents to define plot boundaries. A plot intelligent agent negotiates with neighboring agents based on plot size and POI distributions. The negotiation can be refined by weighting more on either plot size or POI density. All adjustments are operated in a graphical user interface with real-time previews of outcomes so that users can refine segmentation results based on their knowledge of the fields. The final results are saved in text and image files. The text files include plot rows and columns, plot size, and total plot POI. The image files include displays of clusters, POI, and segmented plots. With GRID, users are completely liberated from the labor-intensive task of manually drawing plot lines or polygons. The supervised automation with GRID is expected to enhance the efficiency of agricultural field experiments. 

   more » « less
3. Mental health during the COVID-19 pandemic: Impacts of disease, social isolation, and financial stressors

   https://doi.org/10.1371/journal.pone.0277562  

   Kraut, Robert E.; Li, Han; Zhu, Haiyi (November 2022, PLOS ONE)

   Nabi, Mohammad Hayatun (Ed.)

   BackgroundAlthough research shows that the Covid-19 pandemic has led to declines in mental health, the existing research has not identified the pathways through which this decline happens. AimsThe current study identifies the distinct pathways through which COVID-induced stressors (i.e., social distancing, disease risk, and financial stressors) trigger mental distress and examines the causal impact of these stressors on mental distress. MethodsWe combined evidence of objective pandemic-related stressors collected at the county level (e.g., lack of social contact, infection rates, and unemployment rates) with self-reported survey data from over 11.5 million adult respondents in the United States collected daily for eight months. We used mediation analysis to examine the extent to which the objective stressors influenced mental health by influencing individual respondents’ behavior and fears. ResultsCounty-level, day-to-day social distancing predicted significantly greater mental distress, both directly and indirectly through its effects on individual social contacts, worries about getting ill, and concerns about finances. Economic hardships were indirectly linked to increased mental distress by elevating people’s concerns about their household’s finances. Disease threats were both directly linked to mental distress and indirectly through its effects on individual worries about getting ill. Although one might expect that social distancing from people outside the home would have a greater influence on people who live alone, sub-analyses based on household composition do not support this expectation. ConclusionThis research provides evidence consistent with the thesis that the COVID-19 pandemic harmed the mental well-being of adults in the United States and identifies specific stressors associated with the pandemic that are responsible for increasing mental distress. 

   more » « less
4. Cellular-Assisted COVID-19 Contact Tracing

   https://doi.org/10.1145/3469258.3469848  

   Yi, Fan; Xie, Yaxiong; Jamieson, Kyle (June 2021, Proceedings of the 2nd Workshop on Deep Learning for Wellbeing Applications Leveraging Mobile Devices and Edge Computing (HealthDL))

   null (Ed.)

   The coronavirus disease (COVID-19) pandemic has caused social and economic upheaval around the world. Contact tracing is a proven effective way that health authorities may contain the spread of COVID-19, but is challenging for airborne disease. In this paper, we propose LTESafe, a cellular-assisted privacy-preserving COVID-19 contact tracing system. LTESafe leverages a deep neural network based feature extractor to map the cellular CSI to a high-dimensional feature space, within which the Euclidean distance between points indicates the proximity of devices. By doing so, we preserve user privacy by hiding the physical locations of smartphones and at the same time achieve high accuracy. Our preliminary experimental results demonstrate that LTESafe achieves an overall accuracy of 92.79% in determining whether two devices are within six feet proximity or not, and only misses 1.35% of close contacts. 

   more » « less
5. Time-sensitive POI Recommendation by Tensor Completion with Side Information

   Hui, Bo; Yan, Da; Ku, Wei-Shinn (January 2022, Proceedings of the 38th IEEE International Conference on Data Engineering (ICDE))

   Context has been recognized as an important factor to consider in personalized recommender systems. Particularly in location-based services (LBSs), a fundamental task is to recommend to a mobile user where he/she could be interested to visit next at the right time. Additionally, location-based social networks (LBSNs) allow users to share location-embedded information with friends who often co-occur in the same or nearby points-of-interest (POIs) or share similar POI visiting histories, due to the social homophily theory and Tobler’s first law of geography. So, both the time information and LBSN friendship relations should be utilized for POI recommendation. Tensor completion has recently gained some attention in time-aware recommender systems. The problem decomposes a user-item-time tensor into low-rank embedding matrices of users, items and times using its observed entries, so that the underlying low-rank subspace structure can be tracked to fill the missing entries for time-aware recommendation. However, these tensor completion methods ignore the social-spatial context information available in LBSNs, which is important for POI recommendation since people tend to share their preferences with their friends, and near things are more related than distant things. In this paper, we utilize the side information of social networks and POI locations to enhance the tensor completion model paradigm for more effective time-aware POI recommendation. Specifically, we propose a regularization loss head based on a novel social Hausdorff distance function to optimize the reconstructed tensor. We also quantify the popularity of different POIs with location entropy to prevent very popular POIs from being over-represented hence suppressing the appearance of other more diverse POIs. To address the sensitivity of negative sampling, we train the model on the whole data by treating all unlabeled entries in the observed tensor as negative, and rewriting the loss function in a smart way to reduce the computational cost. Through extensive experiments on real datasets, we demonstrate the superiority of our model over state-of-the-art tensor completion methods. 

   more » « less