Find Research Output

Research Output
  • All
  • Scholar Profiles
  • Research Units
  • Research Output
Filter
Department Publication Year Content Type Data Sources

SELECTED FILTERS

Department of Health and Environmental Sciences
School of Advanced Technology
Clear all

1.Computational modelling of titanium particles in warm spray

Author:Tabbara, H;Gu, S;McCartney, DG

Source:COMPUTERS & FLUIDS,2011,Vol.44

Abstract:A warm spray system has been computationally investigated by introducing a centrally located mixing chamber into a HVOF thermal spray gun. The effects of injecting a cooling gas on the gas and particle dynamics are examined. The gas phase model incorporates liquid fuel droplets which heat, evaporate and then exothermically combust with oxygen within the combustion chamber producing a realistic compressible, supersonic and turbulent jet. The titanium powder is tracked using the Lagrangian approach including particle heating, melting and solidification. The results present an insight into the complex interrelations between the gas and particle phases, and highlight the advantage of warm spray, especially for the deposition of oxygen sensitive materials such as titanium. This work also demonstrates the effectiveness of a computational approach in aiding the development of thermal spray devices. (C) 2011 Elsevier Ltd. All rights reserved.

2.Vehicle Logo Recognition and Attributes Prediction by Multi-task Learning with CNN

Author:Xia, YZ;Feng, J;Zhang, BL

Source:2016 12TH INTERNATIONAL CONFERENCE ON NATURAL COMPUTATION, FUZZY SYSTEMS AND KNOWLEDGE DISCOVERY (ICNC-FSKD),2016,Vol.

Abstract:Vehicle Logo Recognition(VLR) has been an important study field in intelligent Transportation system (ITS). This paper proposes to recognize vehicle logo and predict logo attributes by combining Convolutional Neural Network (CNN) with Multi-Task Learning(MTL). In order to accelerate convergence of multi-task model, an adaptive weight training strategy is employed. To verify the algorithm, the Xiamen University Vehicle logo recognition dataset is extended into a larger vehicle logo dataset including 15 brands, 6 visual attributes and 3 no-visual attributes. The experiment results indicate that the proposed multi-task CNN model perform well for both of logo classification and attribution prediction with overall accuracy 98.14 %%.

3.Urbanization Impacts the Physicochemical Characteristics and Abundance of Fecal Markers and Bacterial Pathogens in Surface Water

Author:Yuan, TM;Vadde, KK;Tonkin, JD;Wang, JJ;Lu, J;Zhang, ZM;Zhang, YX;McCarthy, AJ;Sekar, R

Source:INTERNATIONAL JOURNAL OF ENVIRONMENTAL RESEARCH AND PUBLIC HEALTH,2019,Vol.16

Abstract:Urbanization is increasing worldwide and is happening at a rapid rate in China in line with economic development. Urbanization can lead to major changes in freshwater environments through multiple chemical and microbial contaminants. We assessed the impact of urbanization on physicochemical characteristics and microbial loading in canals in Suzhou, a city that has experienced rapid urbanization in recent decades. Nine sampling locations covering three urban intensity classes (high, medium and low) in Suzhou were selected for field studies and three locations in Huangshan (natural reserve) were included as pristine control locations. Water samples were collected for physicochemical, microbiological and molecular analyses. Compared to medium and low urbanization sites, there were statistically significant higher levels of nutrients and total and thermotolerant coliforms (or fecal coliforms) in highly urbanized locations. The effect of urbanization was also apparent in the abundances of human-associated fecal markers and bacterial pathogens in water samples from highly urbanized locations. These results correlated well with land use types and anthropogenic activities at the sampling sites. The overall results indicate that urbanization negatively impacts water quality, providing high levels of nutrients and a microbial load that includes fecal markers and pathogens.

4.Psychometric Evaluation of Chinese-Language 44-Item and 10-Item Big Five Personality Inventories, Including Correlations with Chronotype, Mindfulness and Mind Wandering

Author:Carciofo, R;Yang, JY;Song, N;Du, F;Zhang, K

Source:PLOS ONE,2016,Vol.11

Abstract:The 44-item and 10-item Big Five Inventory (BFI) personality scales are widely used, but there is a lack of psychometric data for Chinese versions. Eight surveys (total N = 2,496, aged 18-82), assessed a Chinese-language BFI-44 and/or an independently translated Chinese-language BFI-10. Most BFI-44 items loaded strongly or predominantly on the expected dimension, and values of Cronbach's alpha ranged .698-. 807. Test-retest coefficients ranged .694-.770 (BFI-44), and .515-.873 (BFI-10). The BFI-44 and BFI-10 showed good convergent and discriminant correlations, and expected associations with gender (females higher for agreeableness and neuroticism), and age (older age associated with more conscientiousness and agreeableness, and also less neuroticism and openness). Additionally, predicted correlations were found with chronotype (morningness positive with conscientiousness), mindfulness (negative with neuroticism, positive with conscientiousness), and mind wandering/daydreaming frequency (negative with conscientiousness, positive with neuroticism). Exploratory analysis found that the Self-discipline facet of conscientiousness positively correlated with morningness and mindfulness, and negatively correlated with mind wandering/daydreaming frequency. Furthermore, Self-discipline was found to be a mediator in the relationships between chronotype and mindfulness, and chronotype and mind wandering/daydreaming frequency. Overall, the results support the utility of the BFI-44 and BFI-10 for Chinese-language big five personality research.

5.Morningness-eveningness and affect: the mediating roles of sleep quality and metacognitive beliefs

Author:Carciofo, R

Source:SLEEP AND BIOLOGICAL RHYTHMS,2020,Vol.18

Abstract:Morningness (a preference for earlier rise and bed times) is associated with well-being, better sleep quality, and positive affect, while eveningness is associated with poor sleep quality, negative mood, maladaptive behaviours, and psychological disorder. The current study investigated: (1) whether eveningness is associated with more maladaptive metacognitive beliefs; (2) whether maladaptive metacognitive beliefs and sleep quality are related to associations between morningness-eveningness and affect. An online survey with questionnaire measures of morningness-eveningness, affect, sleep quality, and metacognitive beliefs was completed by 591 undergraduate students. More morningness correlated with more positive affect, while more eveningness correlated with more negative affect and poor sleep quality. Eveningness also showed small correlations with having less cognitive confidence and with metacognitive beliefs about uncontrollable thoughts. Mediation analysis showed that cognitive confidence and beliefs about uncontrollable thoughts, together with poor sleep quality (especially subjective sleep quality and daytime dysfunction), were mediators in the relationships between morningness-eveningness and both negative affect and positive affect. These findings suggest that metacognitive beliefs may be an important consideration in understanding the associations between morningness-eveningness and well-being.

6.Relic DNA does not obscure the microbial community of paddy soil microbial fuel cells

Author:Gustave, W;Yuan, ZF;Sekar, R;Toppin, V;Liu, JJY;Ren, YX;Zhang, J;Chen, Z

Source:RESEARCH IN MICROBIOLOGY,2019,Vol.170

Abstract:Soil Microbial Fuel Cells (MFCs) are devices that can generate electricity by using the flooded soil's anode respiring microbial consortium. When the MFC starts to work, the microbial community in the anode vicinity rapidly changes. This shift in the microbial community results in many dead cells that may release their DNA (relic DNA) and obscure culture independent estimates of microbial community composition. Although relic DNA is expected to increase in MFCs, the effect of relic DNA has not been investigated in the soil MFCs system. In this study the effect of the MFCs on the soil microbial community composition within the soil profile and the influence of relic DNA were investigated. Microbial community analysis revealed that the MFCs deployment significantly influenced the community composition within the soil profile. The phylum Proteobacteria (34.4%% vs 23.6%%) and the class Deltaproteobacteria (16.8%% vs 5.9%%) significantly increased in the MFCs compared to the control, while the phylum Firmicutes (24.0%% vs 28.7%%) and the class Sphingobacteria (5.3%% vs 7.0%%) were more abundant in the control. Furthermore, the archaeal phyla Euryarchaeota (40.7%% vs 52.3%%) and Bathyarchaeota (10.1%% vs 17.3%%) were significantly lower in the MFCs, whereas the phylum Woesearchaeota (DHVEG6) (24.4%% vs 19.4%%) was slightly enhanced. Moreover, the results showed that relic DNA can affect the relative abundance of Geobacter and Candidatus Methanoperedens, however, it has no significant effects on the microbial community structure. These results indicate that MFCs can influence the soil microbial community profile, nevertheless the relic DNA generated has minimum effect on the culture independent estimates of microbial community composition. (C) 2018 Institut Pasteur. Published by Elsevier Masson SAS. All rights reserved.

7.Clinical Validity and Reliability of the Malay Language Translations of Gastroesophageal Reflux Disease Questionnaire and Quality of Life in Reflux and Dyspepsia Questionnaire in a Primary Care Setting

Author:Vadivelu, S;Ma, ZF;Ong, EW;Hassan, N;Hassan, NFHN;Aziz, SHSA;Kueh, YC;Lee, YY

Source:DIGESTIVE DISEASES,2019,Vol.37

Abstract:Background: Gastroesophageal Reflux Disease Questionnaire (GERDQ) and Quality of Life in Reflux and Dyspepsia Questionnaire (QOLRAD) are reliable tools for evaluation of GERD. Aim: We aimed to test validity and reliability of Malay language translations of GERDQ and QOLRAD in a primary care setting. Methods: The questionnaires were first translated into the Malay language (GERDQ-M and QOLRAD-M). Patients from primary care clinics with suspected GERD were recruited to complete GERDQ-M, QOLRAD-M, and Malay-translated 36-item short-form health survey (SF-36 or SF36-M), and underwent endoscopy and 24-h pH-impedance test. Results: A total of 104 (mean age 47.1 years, women 51.9%%) participants were enrolled. The sensitivity and specificity for GERDQ-M cut-off score >= 8 were 90.2 and 77.4%%, respectively. Based on this cut-off score, 54.7%% had a high probability of GERD diagnosis. GERD-M score >= 8 vs. <8 was associated with erosive esophagitis (p < 0.001), hiatus hernia (p = 0.03), greater DeMeester score (p = 0.001), and Zerbib scores for acid refluxes (p < 0.001) but not non-acid refluxes (p = 0.1). Mean total scores of QOLRAD-M and SF-36-M were correlated (r = 0.74, p < 0.001). GERDQ-M = 8, erosive esophagitis, and DeMeester >= 14.72 were associated with impaired QOLRAD-M in all domains (all p < 0.02) but this was not seen with SF-36. Conclusions: GERDQ-M and QOLRAD-M are valid and reliable tools applicable in a primary care setting. (C) 2018 S. Karger AG, Basel

8.The effects of the bioanode on the microbial community and element profile in paddy soil

Author:Williamson, G;Chen, Z

Source:ENVIRONMENTAL ARSENIC IN A CHANGING WORLD (AS2018),2018,Vol.

Abstract:In paddy soil the reductive dissolution of iron oxide and the availability of organic matter plays an important role in arsenic release under anaerobic conditions. Microbial fuel cells have been shown to reduce organic matter (OM) content and the rate in which this occurs strongly relate to the external resistance applied. In this study we investigated the effects of bioanode operating at different external resistance on the paddy soil microbial community and iron and arsenic concentration. The results show that MFC can be used to reduce soil pore water iron and arsenic concentration and the extent in which this occurs depend on the external resistance applied. The MFC is able to mitigate arsenic release by decreasing organic matter availability. Furthermore, our finding shows that external resistance had a significant influence on the bacterial community composition that develop on the bioanode however only had minimal effect on the community of the bulk soil. These findings suggest that the sMFC can influence the iron and arsenic concentration by reducing OM content and the microbial community that develop in the bioanode vicinity.

9.Macroinvertebrate drift-benthos trends in a regulated river

Author:Tonkin, JD;Death, RG

Source:FUNDAMENTAL AND APPLIED LIMNOLOGY,2013,Vol.182

Abstract:Downstream drift plays a fundamental role in the spatial distribution and community structure of lotic macroinvertebrates. We sampled both benthic and drifting macroinvertebrates at 15 sites, in three sections of river with varying flow alteration along the Tongariro River, New Zealand. Our objectives were to examine whether (i) benthic and drift density were linearly related throughout the river, (ii) the presence of dams affected the propensity of macroinvertebrates to drift, and (iii) drift propensity was related to benthic periphyton biomass or natural longitudinal patterns down the river. More taxa were collected from the drift than the benthos, although drift and benthic samples were generally taxonomically similar, despite some structural differences. Nonetheless, differences were evident between the major groups when assessing density and relative abundance links between the benthos and drift. The presence of dams did not affect the propensity of macroinvertebrates to drift on the whole, nor was propensity affected by periphyton biomass or distance from source. These results suggest that although altered periphyton biomass in downstream sections in the Tongariro River is altering the composition of benthic and drifting macroinvertebrates, drift propensity is unaffected. However, some deviations from linear relationships between benthic and drift density are evident suggesting these links may be taxon specific.

10.Light attenuation - a more effective basis for the management of fine suspended sediment than mass concentration?

Author:Davies-Colley, RJ;Ballantine, DJ;Elliott, SH;Swales, A;Hughes, AO;Gall, MP

Source:WATER SCIENCE AND TECHNOLOGY,2014,Vol.69

Abstract:Fine sediment continues to be a major diffuse pollution concern with its multiple effects on aquatic ecosystems. Mass concentrations (and loads) of fine sediment are usually measured and modelled, apparently with the assumption that environmental effects of sediment are predictable from mass concentrations. However, some severe impacts of fine sediment may not correlate well with mass concentration, notably those related to light attenuation by suspended particles. Light attenuation per unit mass concentration of suspended particulate matter in waters varies widely with particle size, shape and composition. Data for suspended sediment concentration, turbidity and visual clarity (which is inversely proportional to light beam attenuation) from 77 diverse New Zealand rivers provide valuable insights into the mutual relationships of these quantities. Our analysis of these relationships, both across multiple rivers and within individual rivers, supports the proposition that light attenuation by fine sediment is a more generally meaningful basis for environmental management than sediment mass. Furthermore, optical measurements are considerably more practical, being much cheaper (by about four-fold) to measure than mass concentrations, and amenable to continuous measurement. Mass concentration can be estimated with sufficient precision for many purposes from optical surrogates locally calibrated for particular rivers.

11.Driving Posture Recognition by a Hierarchal Classification System with Multiple Features

Author:Yan, C;Zhang, BL;Coenen, F

Source:2014 7TH INTERNATIONAL CONGRESS ON IMAGE AND SIGNAL PROCESSING (CISP 2014),2014,Vol.

Abstract:This paper presents a novel system for vision-based driving posture recognition. The driving posture dataset was prepared by a side-mounted camera looking at a driver's left profile. After pre-processing for illumination variations, eight action classes of constitutive components of the driving activities were segmented, including normal driving, operating a cell phone, eating and smoking. A global grid-based representation for the action sequence was emphasized, which featured two consecutive steps. Step 1 generates a motion descriptive shape based on a motion frequency image(MFI), and step 2 applies the pyramid histogram of oriented gradients (PHOG) for more discriminating characterization. A three level hierarchal classification system is designed to overcome the difficulties of some overlapping classes. Four commonly applied classifiers, including k-nearest neighbor(KNN), random forest (RF), support vector machine(SVM) and multiple layer perceptron (MLP), are evaluated in each level. The overall classification accuracy is over 87.2%% for the eight classes of driving actions by the proposed classification system.

12.Multi-scale Attention Consistency for Multi-label Image Classification

Author:Xu, Haotian ; Jin, Xiaobo ; Wang, Qiufeng ; Huang, Kaizhu

Source:Communications in Computer and Information Science,2020,Vol.1332

Abstract:Human has well demonstrated its cognitive consistency over image transformations such as flipping and scaling. In order to learn from human’s visual perception consistency, researchers find out that convolutional neural network’s capacity of discernment can be further elevated via forcing the network to concentrate on certain area in the picture in accordance with the human natural visual perception. Attention heatmap, as a supplementary tool to reveal the essential region that the network chooses to focus on, has been developed and widely adopted by CNNs. Based on this regime of visual consistency, we propose a novel end-to-end trainable CNN architecture with multi-scale attention consistency. Specifically, our model takes an original picture and its flipped counterpart as inputs, and then send them into a single standard Resnet with additional attention-enhanced modules to generate a semantically strong attention heatmap. We also compute the distance between multi-scale attention heatmaps of these two pictures and take it as an additional loss to help the network achieve better performance. Our network shows superiority on the multi-label classification task and attains compelling results on the WIDER Attribute Dataset. © 2020, Springer Nature Switzerland AG.

13.Towards Better Forecasting by Fusing Near and Distant Future Visions

Author:Cheng, JZ;Huang, KZ;Zheng, ZB

Source:THIRTY-FOURTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, THE THIRTY-SECOND INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE AND THE TENTH AAAI SYMPOSIUM ON EDUCATIONAL ADVANCES IN ARTIFICIAL INTELLIGENCE,2020,Vol.34

Abstract:Multivariate time series forecasting is an important yet challenging problem in machine learning. Most existing approaches only forecast the series value of one future moment, ignoring the interactions between predictions of future moments with different temporal distance. Such a deficiency probably prevents the model from getting enough information about the future, thus limiting the forecasting accuracy. To address this problem, we propose Multi-Level Construal Neural Network (MLCNN), a novel multi-task deep learning framework. Inspired by the Construal Level Theory of psychology, this model aims to improve the predictive performance by fusing forecasting information (i.e., future visions) of different future time. We first use the Convolution Neural Network to extract multi-level abstract representations of the raw data for near and distant future predictions. We then model the interplay between multiple predictive tasks and fuse their future visions through a modified Encoder-Decoder architecture. Finally, we combine traditional Autoregression model with the neural network to solve the scale insensitive problem. Experiments on three real-world datasets show that our method achieves statistically significant improvements compared to the most state-of-the-art baseline methods, with average 4.59%% reduction on RMSE metric and average 6.87%% reduction on MAE metric.

14.Nitrogen addition increases sexual reproduction and improves seedling growth in the perennial rhizomatous grass Leymus chinensis

Author:Gao, S;Wang, JF;Knops, JMH;Wang, J

Source:BMC PLANT BIOLOGY,2020,Vol.20

Abstract:Background: The Eurasian steppe is an important vegetation type characterized by cold, arid and nitrogen poor conditions. At the Eastern edge, including in the Songnen grassland, the vegetation is dominated by Leymus chinensis (henceforth L. chinensis) and is increasing threatened by elevated anthropogenic nitrogen deposition. L. chinensis is a perennial grass that mainly reproduces vegetatively and its sexual reproduction is limited. However, sexual reproduction plays an important role influencing colonization after large disturbances. To develop an understanding of how elevated nitrogen deposition changes the plant community structure and functioning we need a better understanding how sexual reproduction of L. chinensis changes with nitrogen enrichment. Here we report on a field experiment where we added 10 g N m(-2) yr(-1) and examined changes in seed traits, seed germination and early seedling growth. Results: Nitrogen addition increased seed production by 79%%, contributing to this seed increases were a 28%% increase in flowering plant density, a 40%% increase in seed number per plant and a 11%% increase in seed weight. Seed size increased with a 42%% increase in large seeds and a 49%% decrease in the smallest seed size category. Seed germination success improved by 10%% for small seeds and 18%% for large seeds. Combined, the increased in seed production and improved seed quality doubled the potential seed germination. Subsequent seedling above and below-ground biomass also significantly increased. Conclusions: All aspects of L. chinensis sexual reproduction increased with nitrogen addition. Thus, L. chinensis competitive ability may increase when atmospheric nitrogen deposition increases, which may further reduce overall plant diversity in the low diversity Songnen grasslands.

15.Supernatural Explanatory Models of Health and Illness and HIV Antiretroviral Therapy Use Among Young Men Who Have Sex with Men in China

Author:Pan, SW;Smith, MK;Carpiano, RM;Fu, HY;Ong, JJ;Huang, WT;Tang, WM;Tucker, JD

Source:INTERNATIONAL JOURNAL OF BEHAVIORAL MEDICINE,2020,Vol.27

Abstract:Background In China, men who have sex with men (MSM) shoulder a disproportionate HIV burden. Early initiation and adherence to HIV antiretroviral therapy (ART) will be critical to reversing the HIV epidemic in China, but ART usage remains suboptimal among MSM diagnosed with HIV. One understudied but potentially important factor underpinning suboptimal ART usage is personal belief in supernatural explanatory models of health and illness (supernatural explanatory models). This study examines associations between beliefs in supernatural explanatory models and ART usage among MSM in China. Method In 2017, an online survey was distributed nationwide throughout China by gay community-based organizations. Eligible study participants were self-identified MSM between 16 and 30 years old who had tested positive for HIV and who had seen a doctor in the last 2 years. Beliefs in supernatural explanatory models were measured using a three-item scale developed specifically for the Chinese population (range, 3-15). Results Of 73 participants, the majority were currently using ART (83.6%%) and 42.5%% expressed some endorsement of belief in supernatural explanatory models. However, among 21 participants with the strongest belief in supernatural explanatory models, prevalence of current ART usage was 61.9%%. Stronger belief in supernatural explanatory models was significantly associated with lower likelihood of current ART usage (adjusted odds ratio = 0.52; 95%% confidence interval = 0.13-0.75). Conclusion Belief in supernatural explanatory models may be a powerful predictor of ART usage among MSM living with HIV in China. Further studies are needed to corroborate these findings and elucidate mechanisms of association.

16.Elevated serum uric acid, hyperuricaemia and dietary patterns among adolescents in mainland China

Author:Zhou, H;Ma, ZF;Lu, YM;Du, YY;Shao, J;Wang, LY;Wu, Q;Pan, BY;Zhu, WX;Zhao, QH;Wei, H

Source:JOURNAL OF PEDIATRIC ENDOCRINOLOGY & METABOLISM,2020,Vol.33

Abstract:Background: Elevated serum uric acid concentrations have been associated with metabolic syndrome. However, only limited information is available on the prevalence of hyperuricaemia in adolescents. Therefore, the aim of our cross-sectional study was to study the prevalence of hyperuricaemia and dietary patterns in adolescents aged 13-16 years living in Yangzhou, China. Methods: Adolescents were asked to complete a 20-item food frequency questionnaire (FFQ) and provide an overnight fasting finger-prick sample. Principal component analysis (PCA) with varimax rotation was used to derive the dietary patterns that might be associated with high uric acid concentrations. Results: A total of 1070 adolescents were recruited. Of these, 53.6%% (n = 574) were females, and 58.5%% (n = 625) were within the normal body mass index (BMI) range. The males had a significantly higher serving size and frequency in their weekly food consumption, including meat, poultry, Chinese cereal staple foods and Western-style fast foods, than the females (all p < 0.02). The overall mean serum uric acid concentration and prevalence of hyperuricaemia were 368.6 +/- 114.5 mu mol/L and 37.9%%, respectively. The prevalence of hyperuricaemia was 4.633 times greater among the participants who were overweight and obese than among those who were underweight. On the other hand, the prevalence of hyperuricaemia was 0.694 times lower among the participants who had normal weight than those who were underweight. Conclusions: The prevalence of hyperuricaemia was relatively high in Chinese adolescents. The prevention of hyperuricaemia measures should be strengthened in adolescents to effectively control for obesity and gout, which tend to persist into adulthood.

17.A Covert Ultrasonic Phone-to-Phone Communication Scheme

Author:Shi,Liming;Yu,Limin;Huang,Kaizhu;Zhu,Xu;Wang,Zhi;Li,Xiaofei;Wang,Wenwu;Wang,Xinheng

Source:Lecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering, LNICST,2021,Vol.349

Abstract:© 2021, ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering. Smartphone ownership has increased rapidly over the past decade, and the smartphone has become a popular technological product in modern life. The universal wireless communication scheme on smartphones leverages electromagnetic wave transmission, where the spectrum resource becomes scarce in some scenarios. As a supplement to some face-to-face transmission scenarios, we design an aerial ultrasonic communication scheme. The scheme uses chirp-like signal and BPSK modulation, convolutional code encoding with ID-classified interleaving, and pilot method to estimate room impulse response. Through experiments, the error rate of the ultrasonic communication system designed for mobile phones can be within 0.001 %% in 1 m range. The limitations of this scheme and further research work are discussed as well.

18.Towards an experimental analysis of android phone: GSM network positioning

Author:Man,Ka Lok;Man,Ka Lok;Wang,Wei;Liu,Dawei;Tayahi,Moncef;Hsu,Hui Huang;Lim,Eng Gee

Source:International Journal of Applied Engineering Research,2014,Vol.9

Abstract:Network positioning technology has been widely used in existing smart phones. Traditional network positioning methods are carried out by the network provider; this could violate user's privacy.In this paper we present a work in progress on the positioning of smart phone users using wireless networks. We propose a self-positioning scheme based on the fingerprint method, a positioning method commonly used in indoor environments in previous studies. We outline the proposed self-positioning scheme and propose a k-nearest neighbor method to improve the positioning accuracy. In future, we will evaluate the proposed scheme and method in field experiments.

19.A Systematic Analysis of Link Prediction in Complex Network

Author:Gul, H;Amin, A;Adnan, A;Huang, KZ

Source:IEEE ACCESS,2021,Vol.9

Abstract:Link mining is an important task in the field of data mining and has numerous applications in informal community. Suppose a real-world complex network, the responsibility of this function is to anticipate those links which are not occurred yet in the given real-world network. Holding the significance of LP, the link mining or expectation job has gotten generous consideration from scientists in differing exercise. In this manner, countless strategies for taking care of this issue have been proposed in the late decades. Various articles of link prediction are accessible, however, these are antiquated as multiples new methodologies introduced. In this paper, give a precise assessment of prevail link mining approaches. The investigation is through, it consists the soonest scoring-based approaches and reaches out to the latest strategies which confide on different link prediction strategies. We additionally order link prediction strategies because of their specialized methodology and discussion about the quality and weaknesses of various techniques. Additionally, we compared and expounded various top link prediction techniques. The experimental results of these techniques, over twelve data-sets are ordered here based on performance, RA, 0.7411 > AA, 0.7285 > PA, 0.7202 > Katz, 0.7141 > CN, 0.6951 > HP, 0.6924 > LHN, 0.6017 > PD, 0.3978.

20.Fast graph-based semi-supervised learning and its applications

Author:Zhang,Yan Ming;Huang,Kaizhu;Geng,Guang Gang;Liu,Cheng Lin

Source:Semi-Supervised Learning: Background, Applications and Future Directions,2018,Vol.

Abstract:Despite the great success of graph-based transductive learning methods, most of them have serious problems in scalability and robustness. In this chapter, we propose an efficient and robust graph-based transductive classification method, called minimum tree cut (MTC), which is suitable for large scale data. Motivated from the sparse representation of graph, we approximate a graph by a spanning tree. Exploiting the simple structure, we develop a linear-time algorithm to label the tree such that the cut size of the tree is minimized. This significantly improves graph-based methods, which typically have a polynomial time complexity. Moreover, we theoretically and empirically show that the performance of MTC is robust to the graph construction, overcoming another big problem of traditional graph-based methods. Extensive experiments on public data sets and applications on text extraction fromimages demonstrate our method’s advantages in aspect of accuracy, speed, and robustness.
Total 361 results found
Copyright 2006-2020 © Xi'an Jiaotong-Liverpool University 苏ICP备07016150号-1 京公网安备 11010102002019号