Subjects: Engineering and technical science >> Biomedical Engineering submitted time 2024-04-29
Abstract： The limited training samples pose a significant obstacle to the practical application of brain-computer interface based on error-related potentials (ErrPs), affecting their recognition accuracy. To enhance ErrP recognition under such constraints, we propose a transfer discriminant subspace analysis (TDSA) algorithm that leverages a data alignment strategy. This algorithm extracts a shared discriminant subspace from electroencephalogram samples of both source and target subjects, capturing inter-class differences. By applying temporal alignment within this subspace, it effectively reinforces common features across subjects. We evaluate six different data alignment transfer learning strategies using two publicly available datasets. The results demonstrate that the TDSA algorithm achieves a 6.07% improvement in balanced accuracy for dataset 1 compared to the suboptimal Euclidean alignment method and a 7.88% improvement over non-transfer learning methods. Remarkably, with only 60~100 target subject data samples for training, the TDSA algorithm approaches the classification performance of traditional strategies that require 210~350 samples. This provides a new perspective for facilitating the data alignment of ErrPs.
Peer Review Status:
Awaiting Review
Subjects: Library Science，Information Science >> Library Science submitted time 2023-10-08 Cooperative journals: 《知识管理论坛》
Abstract： [Purpose/significance] Under the background of novel coronavirus pneumonia, this paper proposes a method of identifying COVID-19 news elements in multi-task environment based on transfer learning to provide knowledge services of emergency for the public. [Method/process] Firstly, multiple tasks were used to identify news elements: Time elements were identified based on rules; besides, a cross domain element recognition model was constructed by integrating model transfer and deep learning methods. On this basis, the associated data of COVID-19 news elements was constructed, and the relationship between the elements was displayed by knowledge mapping. [Result/conclusion] The experimental results show that the F1 values of news elements except Drug are above 80%, which indicates that the transfer learning model can achieve fine recognition effect. Moreover, the knowledge map of associated data can intuitively display the key elements and main contents of news. In conclusion, the method proposed in this paper can effectively identify elements in COVID-19 news, thus it can help readers obtain important information from the news accurately and efficiently.
Subjects: Psychology >> Developmental Psychology submitted time 2023-03-28 Cooperative journals: 《心理科学进展》
Abstract： Face pareidolia refers to the compelling illusion of perceiving facial features on inanimate objects, such as an illusory face on the moon surface. Both top-down and bottom-up factors can modulate the occurrence of face pareidolia. In recent years, many studies using behavioral, brain imaging, as well as EEG techniques have been devoted to investigating its influential factors. It was found that the occurrence of face pareidolia depends on whether the stimuli contain face-like structures, whether the internal face template can match the current stimulus, and whether or not there are face related backgrounds. It was also influenced by individual differences and observers’ emotional states. Brain imaging studies suggest that information from the frontal and occipital vision regions can be infused at the fusiform face area (FFA) when experiencing face pareidolia. Future research should focus on exploring the behavioral and neural mechanisms of individual differences in face pareidolia, as well as the interactions and neural mechanisms between different types of top-down modulation.
Subjects: Agriculture, Forestry,Livestock & Aquatic Products Science >> Other Disciplines of Agriculture, Forestry,Livestock & Aquatic Products Science submitted time 2023-02-17 Cooperative journals: 《智慧农业（中英文）》
Abstract： As one of the important way for constructing smart agriculture, unmanned farms are the most attractive in nowadays, and have been explored in many countries. Generally, data, knowledge and intelligent equipment are the core elements of unmanned farms. It deeply integrates modern information technologies such as the Internet of Things, big data, cloud computing, edge computing, and artificial intelligence with agriculture to realize agricultural production information perception, quantitative decision-making, intelligent control, precise input and personalized services. In the paper, the overall technical architecture of unmanned farms is introduced, and five kinds of key technologies of unmanned farms are proposed, which include information perception and intelligent decision-making technology, precision control technology and key equipment for agriculture, automatic driving technology in agriculture, unmanned operation agricultural equipment, management and remote controlling system for unmanned farms. Furthermore, the latest research progress of the above technologies both worldwide are analyzed. Based on which, critical scientific and technological issues to be solved for developing unmanned farms in China are proposed, include unstructured environment perception of farmland, automatic drive for agriculture machinery in complex and changeable farmland environment, autonomous task assignment and path planning of unmanned agricultural machinery, autonomous cooperative operation control of unmanned agricultural machinery group. Those technologies are challenging and absolutely, and would be the most competitive commanding height in the future. The maize unmanned farm constructed in the city of Gongzhuling, Jilin province, China, was also introduced in detail. The unmanned farms is mainly composed of information perception system, unmanned agricultural equipment, management and controlling system. The perception system obtains and provides the farmland information, maize growth, pest and disease information of the farm. The unmanned agricultural machineries could complete the whole process of the maize mechanization under unattended conditions. The management and controlling system includes the basic GIS, remote controlling subsystem, precision operation management subsystem and working display system for unmanned agricultural machineries. The application of the maize unmanned farm has improved maize production efficiency (the harvesting efficiency has been increased by 3－4 times) and reduced labors. Finally, the paper summarizes the important role of the unmanned farm technology were summarized in solving the problems such as reduction of labors, analyzes the opportunities and challenges of developing unmanned farms in China, and put forward the strategic goals and ideas of developing unmanned farm in China.
Subjects: Library Science，Information Science >> Automation method and equipment in intelligence process submitted time 2022-08-31
Abstract： Objective In order to construct a event recognition model in historical books, the performance of sequence labeling method in event recognition in historical ancient books is compared with that of text generation method. Methods In this paper, "Three Kingdoms" is selected as the original corpus. To compare the performance of the two methods, performing on the "Three Kingdoms" event data set, the sequence labeling experiment used BMES annotation and builded the BBCN-SG model ,and the text generation experiment builded the T5-SG model.It also builded RoBERTa-SG and NEZHA-SG models to conduct comparative experiments on generative models. Combining three text generation models and integrating the idea of Stacking ensemble learning, the Stacking-TRN-SG model is constructed. Results On the subject of modeling event recognition in historical ancient books, the performance of the text generation method is significantly better than that of the sequence labeling method. In the text generation method, the performance of the three models is RoBERTa-SG > T5-SG > NEZHA-SG. Stacking ensemble learning greatly improves the recognition performance of generation models. Limitations The computational resources of this paper are limited, and the Stacking-TRN-SG model lacks application research in other historical and ancient corpora. Conclusions The Stacking-TRN-SG model constructed in this paper preliminarily realizes the automatic event recognition of historical ancient books.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2022-05-10 Cooperative journals: 《计算机应用研究》
Abstract： Aiming at the problem that TextRank ignores syntactic information and topic information when extracting chapter keywords, propose a chapter keyword extraction model based on syntactic analysis and topic distribution. Model includes two stages of chapter keyword extraction: paragraph and chapter. Firstly, use paragraphs as a unit to extract paragraph keywords by combining word co-occurrence, grammatical and semantic information. Then cluster the paragraphs according to the paragraph topics to form the paragraph topic set. Finally, extract chapter keywords based on the distribution characteristics of paragraph topics. On the open news dataset, the model's extraction effect improves by about 10% compared with the original TextRank. Results show that the method has significantly improved the extraction effect, and prove the importance of grammatical information and topic information.
Subjects: Biology >> Zoology submitted time 2018-12-24 Cooperative journals: 《动物营养学报》
Abstract：本试验旨在研究湿法挤压膨化加工对棉籽粕中营养物质、游离棉酚含量的影响，以及膨化棉籽粕对生长育肥猪生长性能、血清生化指标及营养物质表观消化率的影响。首先，采用牧羊56×2挤压膨化机和前期优化后的加工参数组合对棉籽粕进行膨化，对比测定棉籽粕和膨化棉籽粕的营养物质和游离棉酚含量的变化。然后，以棉籽粕和膨化棉籽粕为主要试验材料，选取80头体重为（28.78±3.09） kg的杜×长×大三元杂交猪为试验动物，随机分为5个组，每组4个重复，每个重复4头猪（公母各占1/2）。对照组饲喂全玉米–豆粕型基础饲粮，试验1组饲喂添加普通棉籽粕（生长期添加5%普通棉籽粕，育肥期添加10%普通棉籽粕）的饲粮，试验2组、试验3组、试验4组分别饲喂添加膨化棉籽粕（生长期分别添加5%、10%和15%膨化棉籽粕，育肥期分别添加10%、15%和20%膨化棉籽粕）的饲粮，各组饲粮中代谢能和粗蛋白质等营养水平均调配均衡。试验期13周（生长期6周，育肥期7周）。结果表明：1）挤压膨化处理对棉籽粕营养物质含量无明显影响，膨化棉籽粕总氨基酸含量和各个必需氨基酸含量略有升高，游离棉酚含量降低了87.85%。2）生长期，饲粮中添加5%膨化棉籽粕与相同含量的普通棉籽粕相比可提高生长猪的平均日采食量和平均日增重（P>0.05），显著降低料重比（P0.05）。3)育肥期，各膨化棉籽粕组末均重、平均日增重和平均日采食量与对照组和普通棉籽粕组相比差异不显著（P>0.05），但全期试验2组和试验3组料重比显著低于试验1组（P0.05），各膨化棉籽粕组干物质表观消化率均显著高于试验1组（P<0.05）；试验3组和试验4组粗脂肪表观消化率显著高于对照组和试验1组（P<0.05），且随着膨化棉籽粕添加量的增加，粗脂肪表观消化率逐渐升高。饲粮中添加适量的膨化棉籽粕比起添加棉籽粕可显著提高氨基酸表观消化率（P<0.05）。由此可见，挤压膨化加工对棉籽粕营养物质含量影响较小，且能显著降低游离棉酚的含量，在生长育肥猪饲粮中添加膨化棉籽粕可以显著提高生长育肥猪的生长性能、抗氧化能力、免疫能力和营养物质表观消化率，生长猪饲粮中添加量可达15%，育肥猪饲粮中添加量可达20%。
Subjects: Medicine, Pharmacy >> Preclinical Medicine submitted time 2018-01-25 Cooperative journals: 《南方医科大学学报》
Abstract： Objective To explore the role of survivin and PI3K/AKT pathway in the pathogenesis of psoriasis vulgaris (PV). Methods Plaque-like lesions collected from 22 patients with PV in progressive stage and 18 normal control skin specimens were examined using immunohistochemical staining, Western blotting and real-time quantitative PCR for expressions of survivin, PI3K andAKT in the keratinocytes, and their correlation was analyzed.Asmall interfering RNA(siRNA) was used to knock down AKT in cultured HaCaT cells, and Western blotting was used to detect the changes in the expression of survivin. Results Compared with normal skin, PV lesions showed obviously up-regulated expressions of survivin, PI3K andAKT in the keratinocytes. Survivin expression was positively correlated with PI3K (r=0.4510, P=0.0351) andAKT (r=0.4423, P=0.0393) in the keratinocytes in PV lesions. In cultured HaCaT cells, siRNA-mediated knockdown ofAKT caused down-regulation of survivin expression.ConclusionSurvivinandPI3K/AKTsignalingpathwaymayparticipateintheoccurrenceandprogressionofPV.
Subjects: Library Science，Information Science >> Information Science submitted time 2017-11-08 Cooperative journals: 《数据分析与知识发现》
Abstract：【目的】通过实验对比分析, 比较不同停用词表对于不同类型的文本数据的作用效果, 对停用词表的构建与使用提供参考意见。【方法】选取百度停用词表、哈尔滨工业大学停用词表以及四川大学机器智能实验室停用词表, 基于三个不同语料库运用汉语分词技术、TF-IDF 特征评估函数以及VSM 模型进行文本处理, 并且采用Java 编写的K-means 算法进行聚类实验, 通过准确率P、召回率R 和F1 三个评价指标对不同聚类结果进行效果评估。【结果】不同停用词表对于不同类型的文本数据作用效果差异明显, 词表的长度、内容结构是影响作用效果的直接因素, 其中两字停用词作用效果最为明显。【局限】实验文本类型及数量有限, 同时对于不同停用词表仅在词语数量及内容上做了简单的分析比较, 未对停用词按照类别分类进行实验分析。【结论】停用词表对于文本聚类准确度有很大的影响, 构建或选取适宜的中文停用词表极为重要。同时, 过度增加停用词的数量并不会一直改善聚类结果。
Subjects: Library Science，Information Science >> Information Science submitted time 2017-10-11 Cooperative journals: 《数据分析与知识发现》
Abstract： [Objective] Discuss how to obtain the terminology taxonomic relation from Chinese domain unstructured text. [Methods] Based on Digital Library domain text from CNKI, construct terminology hierarchy by terminology extraction, terminology Vector Space Model construction, BIRCH clustering and cluster tag distribution. [Results]Obtain the terminology taxonomic relation of Digital Library domain, and evaluate the effectiveness. The accuracy of clustering reaches up to 80.88%, and the accuracy of cluster tag extraction reaches up to 89.71%. [Limitations]Evaluate the effectiveness by random sampling, and in comparison with one method only. [Conclusions] Making use of BIRCH algorithm to construct terminology taxonomic relation, this algorithm has obvious advantage compared with K-means clustering method, and has higher execution and clustering effectiveness.
Subjects: Library Science，Information Science >> Information Science submitted time 2017-10-11 Cooperative journals: 《数据分析与知识发现》
Abstract：【目的】探讨冶金领域中文专利术语抽取模型的最优条件, 用于有效地抽取冶金领域专利术语。【方法】使用尚不完善的核心语料库, 在无需人工标引的情况下, 采用条件随机场(CRFs)构建字角色标注的冶金领域中文专利术语识别模型。详细说明模型的构建过程, 同时重点对比CFRs 的各个因素(特征组合、字长窗口等)对识别效果的影响。【结果】实验结果表明字序列、级别特征、领域特征、温度特征的组合在字长窗口为3, c 等于1,f 等于1 时, 准确率达到94.26%, 召回率达到94.37%, F1 值达到94.5%。【局限】核心词典欠完善, 使得部分词语标注不够准确; 未与其他方法作详细比较, 未详细说明CRFs 的可靠性。【结论】CRFs 在适当的角色和特征以及特征模板的组合下能较好地识别出冶金领域的中文专利术语。
Subjects: Materials Science >> Materials Science (General) submitted time 2017-04-10 Cooperative journals: 《材料研究学报》
Abstract：用膨胀型阻燃剂(IFR)和乙烯辛烯共聚物(POE)对聚丙烯(iPP)进行阻燃和增韧改性,比较研究了两种典型增容剂聚丙烯接枝马来酸酐(PP-g-MAH)和乙烯辛烯共聚物接枝马来酸酐(POE-g-MAH)对膨胀阻燃增韧共混复合体系阻燃性能以及力学性能的影响。结果表明：IFR可提高聚丙烯共混物的燃烧性能,但是明显降低材料的力学性能,而增容剂的加入可同时提高复合材料的燃烧性能和力学性能。PP-g-MAH使IFR的分散更均匀,添加1%(质量分数,下同)的PP-g-MAH使复合材料的平均热释放速率、热释放速率峰值、比消光面积平均值以及烟释放总量比未添加增容剂的阻燃材料分别下降24%、30%、56%和46%;而POE-g-MAH能使复合材料形成包覆结构,使其冲击强度明显提高,加入5%的POE-g-MAH可使复合材料冲击强度提高93%。
Hits
Hits
Downloads
Comment
YES