47 research outputs found
Distributed frameworks for approximate data analytics
No full textData-driven discovery has become critical to the mission of many enterprises and scientific research. At the same time, the rate of data production and collection is outpacing technology scaling, suggesting that significant future investment, time, and energy will be needed for data processing. Straightforwardly increasing hardware resources can address the extra processing needs by either adding more CPU cores/memory (scale-up) or more worker nodes (scale-out). However, it will introduce higher computing cost that may not be feasible when budget is limited. One powerful tool to address the above challenge is approximate computing, which trades off computational time and resources with computational accuracy by reducing the amount of data needed to be processed. Fortunately, many data analytic applications such as data mining, log processing, video/image processing are amenable to approximation.
In this thesis, we describe the design and implementation of approximation frameworks to accelerate distributed data analytics. We present the frameworks targeting a variety of tasks and datasets, including log aggregation, text analytics and video querying and aggregation:
Our first work targets approximating aggregation jobs with error estimation. Aggregation is central to many decision support queries. Aggregation is also an important component in OLAP~(Online Analytical Processing) systems, and is frequently used for summarizing data patterns in business intelligence. Aggregation jobs often involve multiple transformation steps in a data processing pipeline. We design and implement a sampling-based approximation framework called ApproxSpark, that can rigorously derive estimators with error bounds for approximate aggregation.
Our second work targets approximate text analytic tasks. We propose and evaluate a framework called EmApprox that uses sampling-based approximation to speed up the processing of a wide range of queries over large text datasets. EmApprox builds an index for a dataset by learning a natural language processing model, producing vectors representing words and subcollections of documents. Our approximation index can significantly improve approximate quality while processing a small amount of the data. It will apply to each sampling unit with a sampling rate proportional to its similarity to the query. We have implemented a prototype of EmApprox as a Python library, and used it to approximate aggregation, information retrieval, and recommendation tasks.
Finally, we target approximate video analytics. Video data embed rich and high-quality information. Yet video analytics is particularly compute intensive as it often involves invoking a deep convolutional neural network~(CNN) for object detection. We design and implement a approximate video analytics framework called VidApprox for accelerating video queries that involve object detection. VidApprox first leverages cheap CNNs to learn vector representations of video segments, and further processes the vectors as a persistent index structure. At query processing time, the index lookup will serve as auxiliary information for only retrieving a subset of more similar video segments. It make downstream processing such as object detection or aggregation more efficient by only performing expensive operations such as CNN inference on the relevant video data.
We show that approximation is a promising technique for reducing processing time for large datasets. However, approximation poses multifaceted challenges when applied to data processing tasks across different domains. In particular, approximation when applied can present a complicated trade-off space that involves processing time reduction, quality of computation results and preprocessing complexity. Our works not only demonstrates that it is possible to balance computational accuracy with processing time reduction, but also that a machine-learned compact representation of the data generated can function as index structure for improving approximation quality across different domains and data sets.Ph.D.Includes bibliographical reference
The research and application of electricity economic boom model under the constraint of energy conservation and emission reduction
No full textDifferent Effects of Nano-Crystallization on High-Temperature Oxidation Behavior of Cu–Si and Fe–Si Alloys
No full textHow AI and Robotics Will Advance Interventional Radiology: Narrative Review and Future Perspectives
No full textThe rapid advancement of artificial intelligence (AI) and robotics has led to significant progress in various medical fields including interventional radiology (IR). This review focuses on the research progress and applications of AI and robotics in IR, including deep learning (DL), machine learning (ML), and convolutional neural networks (CNNs) across specialties such as oncology, neurology, and cardiology, aiming to explore potential directions in future interventional treatments. To ensure the breadth and depth of this review, we implemented a systematic literature search strategy, selecting research published within the last five years. We conducted searches in databases such as PubMed and Google Scholar to find relevant literature. Special emphasis was placed on selecting large-scale studies to ensure the comprehensiveness and reliability of the results. This review summarizes the latest research directions and developments, ultimately analyzing their corresponding potential and limitations. It furnishes essential information and insights for researchers, clinicians, and policymakers, potentially propelling advancements and innovations within the domains of AI and IR. Finally, our findings indicate that although AI and robotics technologies are not yet widely applied in clinical settings, they are evolving across multiple aspects and are expected to significantly improve the processes and efficacy of interventional treatments
Impact of adult sons’ marriage squeeze on the mental health of parents in rural China
No full textABSTRACT: Studying the influencing factors and interventions of mental health in rural China has important practical significance for promoting personal development and social progress. Using the 2016 waves of the China Family Panel Studies data, this paper examined the effect of adult sons’ marriage squeeze on parents’ mental health in rural China and its corresponding mechanisms. Our study showed a significant negative association between adult sons’ marriage squeeze and the mental health status of rural parents. Moreover, the negative association of adult sons’ marriage squeeze with mental health is larger for female parents, parents who are less than 60 years old and parents living with children. In addition, we investigated possible mechanisms of the effect, including intergenerational support, intergenerational relationship and neighborhood relationship, and that adult sons’ marriage squeeze mainly damages the mental health of rural parents by increasing the downstream intergenerational support of rural parents and worsening intergenerational and neighborhood relationships.</div
Multi-Slice CT Features Predict Pathological Risk Classification in Gastric Stromal Tumors Larger Than 2 cm: A Retrospective Study
No full textBackground: The Armed Forces Institute of Pathology (AFIP) had higher accuracy and reliability in prognostic assessment and treatment strategies for patients with gastric stromal tumors (GSTs). The AFIP classification is frequently used in clinical applications. But the risk classification is only available for patients who are previously untreated and received complete resection. We aimed to investigate the feasibility of multi-slice MSCT features of GSTs in predicting AFIP risk classification preoperatively. Methods: The clinical data and MSCT features of 424 patients with solitary GSTs were retrospectively reviewed. According to pathological AFIP risk criteria, 424 GSTs were divided into a low-risk group (n = 282), a moderate-risk group (n = 72), and a high-risk group (n = 70). The clinical data and MSCT features of GSTs were compared among the three groups. Those variables (p < 0.05) in the univariate analysis were included in the multivariate analysis. The nomogram was created using the rms package. Results: We found significant differences in the tumor location, morphology, necrosis, ulceration, growth pattern, feeding artery, vascular-like enhancement, fat-positive signs around GSTs, CT value in the venous phase, CT value increment in the venous phase, longest diameter, and maximum short diameter (all p < 0.05). Two nomogram models were successfully constructed to predict the risk of GSTs. Low- vs. high-risk group: the independent risk factors of high-risk GSTs included the location, ulceration, and longest diameter. The area under the receiver operating characteristic curve (AUC) of the prediction model was 0.911 (95% CI: 0.872–0.951), and the sensitivity and specificity were 80.0% and 89.0%, respectively. Moderate- vs. high-risk group: the morphology, necrosis, and feeding artery were independent risk factors of a high risk of GSTs, with an AUC value of 0.826 (95% CI: 0.759–0.893), and the sensitivity and specificity were 85.7% and 70.8%, respectively. Conclusions: The MSCT features of GSTs and the nomogram model have great practical value in predicting pathological AFIP risk classification between high-risk and non-high-risk groups before surgery
