Academic literature on the topic 'Conditional-based monitoring'

Natural language processing (NLP) technology has played a pivotal role in health monitoring as an important artificial intelligence method. As a key technology in NLP, relation triplet extraction is closely related to the performance of health monitoring. In this paper, a novel model is proposed for joint extraction of entities and relations, combining conditional layer normalization with the talking-head attention mechanism to strengthen the interaction between entity recognition and relation extraction. In addition, the proposed model utilizes position information to enhance the extraction accuracy of overlapping triplets. Experiments on the Baidu2019 and CHIP2020 datasets demonstrate that the proposed model can effectively extract overlapping triplets, which leads to significant performance improvements compared with baselines.

The pedometer, an objective assessment of measuring step counts, has often been used to motivate individuals to increase their ambulatory physical activity. Minimal contact pedometer-based intervention (MCPBI) is gaining in popularity because they are simple and inexpensive. MCPBI is based on self-monitoring by the participants; however, one limitation of using the self-monitoring approach was the participant attrition (i.e., dropout), which makes it difficult to achieve the successful intervention. A new algorithm for pedometer-based intervention, the systematic-monitoring based on conditional feedback, was designed to increase awareness and allow participants to more successfully attain their step goals. Thus, the purpose of this study was to examine the effect of the systematic-monitoring based on conditional feedback algorithm on 10,000 step goal attainments. The study result can be used to design more comprehensive pedometer-based physical activity interventions to increase individuals’ overall health status.

Conditional learning is an important component of our everyday activities (e.g., handling a phone or sorting work files) and requires identification of the arbitrary stimulus, accurate selection of the motor response, monitoring of the response, and storing in memory of the stimulus-response association for future recall. Learning this type of conditional visuomotor task appears to engage the premotor dorsal region (PMd). However, the extent to which PMd might be involved in specific or all processes of conditional learning is not well understood. Using transcranial magnetic stimulation (TMS), we demonstrate the role of human PMd in specific stages of learning of a novel conditional visuomotor task that required subjects to identify object center of mass using a color cue and to apply appropriate torque on the object at lift onset to minimize tilt. TMS over PMd, but not vertex, increased error in torque exerted on the object during the learning trials. Analyses of digit position and forces further revealed that the slowing in conditional visuomotor learning resulted from impaired monitoring of the object orientation during lift, rather than stimulus identification, thus compromising the ability to accurately reduce performance error across trials. Importantly, TMS over PMd did not alter production of torque based on the recall of learned color-torque associations. We conclude that the role of PMd for conditional learning is highly sensitive to the stage of learning visuomotor associations. NEW & NOTEWORTHY Conditional learning involves stimulus identification, motor response selection, response monitoring, memory encoding, and recall of the learned association. Premotor dorsal (PMd) has been implicated for conditional learning. However, the extent to which PMd might be involved in specific or all stages of conditional learning is not well understood. The novel finding of our study is that PMd appears to be involved with monitoring motor responses, a sensorimotor integration stage essential for conditional learning.

In a real interactive service system, a smart meter can only read the total amount of energy consumption rather than analyze the internal load components for users. Nonintrusive load monitoring (NILM), as a vital part of smart power utilization techniques, can provide load disaggregation information, which can be further used for optimal energy use. In our paper, we introduce a new method called linear-chain conditional random fields (CRFs) for NILM and combine two promising features: current signals and real power measurements. The proposed method relaxes the independent assumption and avoids the label bias problem. Case studies on two open datasets showed that the proposed method can efficiently identify multistate appliances and detect appliances that are not easily identified by other models.