Search results for: temporal data

Authors: Eui-Jin Kim, Hasik Lee, Su-Jin Park, Dong-Kyu Kim

Abstract:

Smart-card data are expected to provide information on activity pattern as an alternative to conventional person trip surveys. The focus of this study is to propose a method for training the person trip surveys to supplement the smart-card data that does not contain the purpose of each trip. We selected only available features from smart card data such as spatiotemporal information on the trip and geographic information system (GIS) data near the stations to train the survey data. XGboost, which is state-of-the-art tree-based ensemble classifier, was used to train data from multiple sources. This classifier uses a more regularized model formalization to control the over-fitting and show very fast execution time with well-performance. The validation results showed that proposed method efficiently estimated the trip purpose. GIS data of station and duration of stay at the destination were significant features in modeling trip purpose.

Keywords: activity pattern, data fusion, smart-card, XGboost

Procedia PDF Downloads 246

Authors: A. Rong, P. B. Luh, R. Lahdelma

Abstract:

High penetration of intermittent renewable energy sources (RES) such as solar power and wind power into the energy system has caused temporal and spatial imbalance between electric power supply and demand for some countries and regions. This brings about the critical need for coordinating power production and power exchange for different regions. As compared with the power-only systems, the combined heat and power (CHP) systems can provide additional flexibility of utilizing RES by exploiting the interdependence of power and heat production in the CHP plant. In the CHP system, power production can be influenced by adjusting heat production level and electric power can be used to satisfy heat demand by electric boiler or heat pump in conjunction with heat storage, which is much cheaper than electric storage. This paper addresses multi-site CHP systems without considering RES, which lay foundation for handling penetration of RES. The problem under study is the unit commitment (UC) of the transmission-constrained multi-site CHP systems. We solve the problem by combining linear relaxation of ON/OFF states and sequential dynamic programming (DP) techniques, where relaxed states are used to reduce the dimension of the UC problem and DP for improving the solution quality. Numerical results for daily scheduling with realistic models and data show that DP-based algorithm is from a few to a few hundred times faster than CPLEX (standard commercial optimization software) with good solution accuracy (less than 1% relative gap from the optimal solution on the average).

Keywords: dynamic programming, multi-site combined heat and power system, relaxed states, transmission-constrained generation unit commitment

Procedia PDF Downloads 365

Authors: Haojie Wang, Xun Li, Rui Yin

Abstract:

In order to solve the memorization overfitting in the model-agnostic meta-learning MAML algorithm, a method of generating mutually exclusive tasks based on data augmentation is proposed. This method generates a mutex task by corresponding one feature of the data to multiple labels so that the generated mutex task is inconsistent with the data distribution in the initial dataset. Because generating mutex tasks for all data will produce a large number of invalid data and, in the worst case, lead to an exponential growth of computation, this paper also proposes a key data extraction method that only extract part of the data to generate the mutex task. The experiments show that the method of generating mutually exclusive tasks can effectively solve the memorization overfitting in the meta-learning MAML algorithm.

Keywords: mutex task generation, data augmentation, meta-learning, text classification.

Procedia PDF Downloads 143

Authors: Sukhwinder Singh, Julian Park, Dinesh Kumar Benbi

Abstract:

Although soil health, which is recognized as one of the key determinants of sustainable agricultural development, can be measured by a range of physical, chemical and biological parameters, the widely used parameters include pH, electrical conductivity (EC), organic carbon (OC), plant available phosphorus (P) and potassium (K). Soil health is largely affected by the occurrence of natural events or human activities and can be improved by various land management practices. A database of 120 soil samples collected from farmers’ fields spread across three major agro-climatic zones of Punjab suggested that the average pH, EC, OC, P and K was 8.2 (SD = 0.75, Min = 5.5, Max = 9.1), 0.27 dS/m (SD = 0.17, Min = 0.072 dS/m, Max = 1.22 dS/m), 0.49% (SD = 0.20, Min = 0.06%, Max = 1.2%), 19 mg/kg soil (SD = 22.07, Min = 3 mg/kg soil, Max = 207 mg/kg soil) and 171 mg/kg soil (SD = 47.57, Min = 54 mg/kg soil, Max = 288 mg/kg soil), respectively. Region-wise, pH, EC and K were the highest in south-western district of Ferozpur whereas farmers in north-eastern district of Gurdaspur had the best soils in terms of OC and P. The soils in the central district of Barnala had lower OC, P and K than the respective overall averages while its soils were normal but skewed towards alkalinity. Besides agro-climatic conditions, the size of landholding and farmer education showed a significant association with Soil Fertility Index (SFI), a composite index calculated using the aforementioned parameters’ normalized weightage. All the four stakeholder groups cited the current cropping patterns, burning of rice crop residue, and imbalanced use of chemical fertilizers for change in soil health. However, the current state of soil health in Punjab is unclear, which needs further investigation based on temporal data collected from the same field to see the short and long-term impacts of various crop combinations and varied cropping intensity levels on soil health.

Keywords: soil health, punjab agriculture, sustainability, soil fertility index

Procedia PDF Downloads 362

Authors: Mohammadamin Hashemi, Mohammadreza Kashizadeh

Abstract:

Droughts pose a significant threat to sustainable water resource management, agriculture, and socioeconomic sectors, particularly in the field of climate change. This study investigates drought simulation using rainfall-runoff modelling in the Cuneo district, Italy, over the past 60-year period. The study leverages the TUW model, a lumped conceptual rainfall-runoff model with a semi-distributed operation capability. Similar in structure to the widely used Hydrologiska Byråns Vattenbalansavdelning (HBV) model, the TUW model operates on daily timesteps for input and output data specific to each catchment. It incorporates essential routines for snow accumulation and melting, soil moisture storage, and streamflow generation. Multiple catchments' discharge data within the Cuneo district form the basis for thorough model calibration employing the Kling-Gupta Efficiency (KGE) metric. A crucial metric for reliable drought analysis is one that can accurately represent low-flow events during drought periods. This ensures that the model provides a realistic picture of water availability during these critical times. Subsequent validation of monthly discharge simulations thoroughly evaluates overall model performance. Beyond model development, the investigation delves into drought analysis using the robust Standardized Runoff Index (SRI). This index allows for precise characterization of drought occurrences within the study area. A meticulous comparison of observed and simulated discharge data is conducted, with particular focus on low-flow events that characterize droughts. Additionally, the study explores the complex interplay between land characteristics (e.g., soil type, vegetation cover) and climate variables (e.g., precipitation, temperature) that influence the severity and duration of hydrological droughts. The study's findings demonstrate successful calibration of the TUW model across most catchments, achieving commendable model efficiency. Comparative analysis between simulated and observed discharge data reveals significant agreement, especially during critical low-flow periods. This agreement is further supported by the Pareto coefficient, a statistical measure of goodness-of-fit. The drought analysis provides critical insights into the duration, intensity, and severity of drought events within the Cuneo district. This newfound understanding of spatial and temporal drought dynamics offers valuable information for water resource management strategies and drought mitigation efforts. This research deepens our understanding of drought dynamics in the Cuneo region. Future research directions include refining hydrological modelling techniques and exploring future drought projections under various climate change scenarios.

Keywords: hydrologic extremes, hydrological drought, hydrological modelling, machine learning, rainfall-runoff modelling

Procedia PDF Downloads 41

Authors: Milind Chaudhari, Suhail Balasinor

Abstract:

Due to massive deforestation and an ever-increasing population, the organic content of the soil is depleting at a much faster rate. Due to this, there is a big chance that the entire food production in the world will drop by 40% in the next two decades. Vertical farming can help in aiding food production by leveraging big data and cloud computing to ensure plants are grown naturally by providing the optimum nutrients sunlight by analyzing millions of data points. This paper outlines the most important parameters in vertical farming and how a combination of big data and AI helps in calculating and analyzing these millions of data points. Finally, the paper outlines how different organizations are controlling the indoor environment by leveraging big data in enhancing food quantity and quality.

Keywords: big data, IoT, vertical farming, indoor farming

Procedia PDF Downloads 175

Authors: M. E. Hammond, R. Pokorny

Abstract:

Not much research has gone into the dynamics of natural regeneration of trees species in tropical forest regions. This study seeks to investigate the impact of gap sizes and light distribution in forest floors on the regeneration of Celtis mildbraedii (CEM), Nesogordonia papaverine (NES) and Terminalia superba (TES). These are selected economically important tree species with different shade tolerance attributes. The spatial distribution patterns and the potential regeneration competition index (RCI) among species using height to diameter ratio (HDR) have been assessed. Gap sizes ranging between 287 – 971 m² were selected at the Bia Tano forest reserve, a tropical moist semi-deciduous forest in Ghana. Four (4) transects in the cardinal directions were constructed from the center of each gap. Along each transect, ten 1 m² sampling zones at 2 m spacing were established. Then, three gap microsites (labeled ecozones I, II, III) were delineated within these sampling zones based on the varying temporal light distribution on the forest floor. Data on height (H), root collar diameter (RCD) and regeneration census were gathered from each of the ten sampling zones. CEM and NES seedlings (≤ 50 cm) and saplings (≥ 51 cm) were present in all ecozones of the large gaps. Seedlings of TES were observed in all ecozones of large and small gaps. Regression analysis showed a significant negative linear relationship between independent RCD and H growth variables on dependent HDR index in ecozones II and III of both large and small gaps. There was a correlation between RCD and H in both large and small gaps. A strong regeneration competition was observed among species in ecozone II in large (df 2, F=3.6, p=0.035) and small (df 2, F=17.9, p=0.000) gaps. These results contribute to the understanding of the natural regeneration of different species with regards to light regimes in forest floors.

Keywords: Celtis mildbraedii, ecozones, gaps, Nesogordonia papaverifera, regeneration, Terminalia superba

Procedia PDF Downloads 139

Authors: Christakis Damianou, Christos Christofi, Nicos Mylonas

Abstract:

The aim of this paper was to conduct a feasibility study using a flat rectangular (3x10 mm2) MRI compatible transducer operating at 5 MHz for destroying atherosclerotic plaque using the thermal effects of ultrasound in in vitro models. A parametric study was performed where the time needed to ablate the plaque was studied as a function of Spatial Average Temporal Average (SATA) intensity, and pulse duration. The time needed to ablate plaque is directly related to intensity, and pulse duration. The temperature measured close to the artery is above safe limits and therefore thermal ultrasound does not have a place in removing plaques in arteries.

Keywords: ultrasound, atherosclerotic, plaque, pulse

Procedia PDF Downloads 293

Authors: A. Anis, W. Bekheet, A. El Hakim

Abstract:

Implementing a Road Safety Management System (SMS) in a crowded developing country such as Egypt is a necessity. Beginning a sustainable SMS requires a comprehensive reliable data system for all information pertinent to road crashes. In this paper, a survey for the available data in Egypt and validating it for using in an SMS in Egypt. The research provides some missing data, and refer to the unavailable data in Egypt, looking forward to the contribution of the scientific society, the authorities, and the public in solving the problem of missing or unreliable crash data. The required data for implementing an SMS in Egypt are divided into three categories; the first is available data such as fatality and injury rates and it is proven in this research that it may be inconsistent and unreliable, the second category of data is not available, but it may be estimated, an example of estimating vehicle cost is available in this research, the third is not available and can be measured case by case such as the functional and geometric properties of a facility. Some inquiries are provided in this research for the scientific society, such as how to improve the links among stakeholders of road safety in order to obtain a consistent, non-biased, and reliable data system.

Keywords: road safety management system, road crash, road fatality, road injury

Procedia PDF Downloads 146

Authors: Oualid Walid Ben Ali

Abstract:

Big Data has become one of the buzzwords today. The recent explosion of digital data has led the organization, either private or public, to a new era towards a more efficient decision making. At some point, business decided to use that concept in order to learn what make their clients tick with phrases like ‘sales funnel’ analysis, ‘actionable insights’, and ‘positive business impact’. So, it stands to reason that Big Data was viewed through green (read: money) colored lenses. Somewhere along the line, however someone realized that collecting and processing data doesn’t have to be for business purpose only, but also could be used for other purposes to assist law enforcement or to improve policing or in road safety. This paper presents briefly, how Big Data have been used in the fields of policing order to improve the decision making process in the daily operation of the police. As example, we present a big-data driven system which is sued to accurately dispatch the patrol cars in a geographic environment. The system is also used to allocate, in real-time, the nearest patrol car to the location of an incident. This system has been implemented and applied in the Emirate of Abu Dhabi in the UAE.

Keywords: big data, big data analytics, patrol car allocation, dispatching, GIS, intelligent, Abu Dhabi, police, UAE

Procedia PDF Downloads 490

Authors: Tamar Trop, Boris Portnov

Abstract:

Public Space Lighting (PSL) of outdoor urban areas promotes comfort, defines spaces and neighborhood identities, enhances perceived safety and security, and contributes to residential satisfaction and wellbeing. However, if excessive or misdirected, PSL leads to unnecessary energy waste and increased greenhouse gas emissions, poses a non-negligible threat to the nocturnal environment, and may become a potential health hazard. At present, PSL is designed according to international, regional, and national standards, which consolidate best practice. Yet, knowledge regarding the optimal light characteristics needed for creating a perception of personal comfort and safety in densely populated residential areas, and the factors associated with this perception, is still scarce. The presented study suggests a paradigm shift in designing PSL towards a user-centered approach, which incorporates pedestrians' perspectives into the process. The study is an ongoing joint research project between China and Israel Ministries of Science and Technology. Its main objectives are to reveal inhabitants' perceptions of and preferences for PSL in different densely populated neighborhoods in China and Israel, and to develop a model that links instrumentally measured parameters of PSL (e.g., intensity, spectra and glare) with its perceived comfort and quality, while controlling for three groups of attributes: locational, temporal, and individual. To investigate measured and perceived PSL, the study employed various research methods and data collection tools, developed a location-based mobile application, and used multiple data sources, such as satellite multi-spectral night-time light imagery, census statistics, and detailed planning schemes. One of the study’s preliminary findings is that higher sense of safety in the investigated neighborhoods is not associated with higher levels of light intensity. This implies potential for energy saving in brightly illuminated residential areas. Study findings might contribute to the design of a smart and adaptive PSL strategy that enhances pedestrians’ perceived safety and comfort while reducing light pollution and energy consumption.

Keywords: energy efficiency, light pollution, public space lighting, PSL, safety perceptions

Procedia PDF Downloads 133

Authors: Shah Fahad

Abstract:

Managing the Urban Heat Island (UHI) effects is a pressing concern for achieving sustainable urban development and ensuring thermal comfort in major cities of developing nations, such as Lahore, Pakistan. The current UHI effect is mostly triggered by climate change and rapid urbanization. This study explored UHI over the Lahore district and its adjoining urban and rural-urban fringe areas. Landsat satellite data was utilized to investigate spatiotemporal patterns of Land Use and Land Cover changes (LULC), Land Surface Temperature (LST), UHI, Normalized Difference Built-up Index (NDBI), Normalized Difference Vegetation Index (NDVI), and Urban Thermal Field Variance Index (UTFVI). The built-up area increased very fast, with a coverage of 22.99% in 2000, 36.06% in 2010, and 47.17% in 2020, while vegetation covered 53.21 % in 2000 and 46.16 % in 2020. It also revealed a significant increase in the mean LST, from 33°C in 2000 to 34.8°C in 2020. The results indicated a significantly positive correlation between LST and NDBI, a weak correlation was also observed between LST and NDVI. The study used scatterplots to show the correlation between NDBI and NDVI with LST, results revealed that the NDBI and LST had an R² value of 0.6831 in 2000 and 0.06541 in 2022, while NDVI and LST had an R² value of 0.0235 in 1998 and 0.0295 in 2022. Proper environmental planning is vital in specific locations to enhance quality of life, protect the ecosystem, and mitigate climate change impacts.

Keywords: land use land cover, spatio-temporal analysis, remote sensing, land surface temperature, urban heat island, lahore pakistan

Procedia PDF Downloads 77

Authors: Xu Du, Aparna S. Varde

Abstract:

In this research, we propose to conduct diagnostic and predictive analysis about the key factors and consequences of urban population relocation. To achieve this goal, urban simulation models extract the urban development trends as land use change patterns from a variety of data sources. The results are treated as part of urban big data with other information such as population change and economic conditions. Multiple data mining methods are deployed on this data to analyze nonlinear relationships between parameters. The result determines the driving force of population relocation with respect to urban sprawl and urban sustainability and their related parameters. Experiments so far reveal that data mining methods discover useful knowledge from the multicity urban data. This work sets the stage for developing a comprehensive urban simulation model for catering to specific questions by targeted users. It contributes towards achieving sustainability as a whole.

Keywords: data mining, environmental modeling, sustainability, urban planning

Procedia PDF Downloads 308

Authors: Aref Ghafouri, Mohammad javad Mollakazemi, Farhad Asadi

Abstract:

In this paper, model order reduction method is used for approximation in linear and nonlinearity aspects in some experimental data. This method can be used for obtaining offline reduced model for approximation of experimental data and can produce and follow the data and order of system and also it can match to experimental data in some frequency ratios. In this study, the method is compared in different experimental data and influence of choosing of order of the model reduction for obtaining the best and sufficient matching condition for following the data is investigated in format of imaginary and reality part of the frequency response curve and finally the effect and important parameter of number of order reduction in nonlinear experimental data is explained further.

Keywords: frequency response, order of model reduction, frequency matching condition, nonlinear experimental data

Procedia PDF Downloads 402

Authors: Thuan Nguyen

Abstract:

In the present time, data to a data-driven knowledge-based economy is the same as oil to the industrial age hundreds of years ago. Data is everywhere in vast volumes! Big data analytics is expected to help firms not only efficiently improve performance but also completely transform how they should run their business. However, employing the emergent technology successfully is not easy, and assessing the roles of big data in improving firm performance is even much harder. There was a lack of studies that have examined the impacts of big data analytics on organizational performance. This study aimed to fill the gap. The present study suggested using firms’ intellectual capital as a proxy for big data in evaluating its impact on organizational performance. The present study employed the Value Added Intellectual Coefficient method to measure firm intellectual capital, via its three main components: human capital efficiency, structural capital efficiency, and capital employed efficiency, and then used the structural equation modeling technique to model the data and test the models. The financial fundamental and market data of 100 randomly selected publicly listed firms were collected. The results of the tests showed that only human capital efficiency had a significant positive impact on firm profitability, which highlighted the prominent human role in the impact of big data technology.

Keywords: big data, big data analytics, intellectual capital, organizational performance, value added intellectual coefficient

Procedia PDF Downloads 245

Authors: Hitesh Tahbildar

Abstract:

The cost of test data generation for a program is computationally very high. In general case, no algorithm to generate test data for all types of algorithms has been found. The cost of generating test data for different types of algorithm is different. Till date, people are emphasizing the need to generate test data for different types of programming constructs rather than different types of algorithms. The test data generation methods have been implemented to find heuristics for different types of algorithms. Some algorithms that includes divide and conquer, backtracking, greedy approach, dynamic programming to find the minimum cost of test data generation have been tested. Our experimental results say that some of these types of algorithm can be used as a necessary condition for selecting heuristics and programming constructs are sufficient condition for selecting our heuristics. Finally we recommend the different heuristics for test data generation to be selected for different types of algorithms.

Keywords: ongest path, saturation point, lmax, kL, kS

Procedia PDF Downloads 405

Authors: Hatice Kübra Koç

Abstract:

For academia, collecting reliable and valid data is one of the most significant issues for researchers. However, it is not the same procedure for all different target groups; meanwhile, during data collection from teenagers, young adults, or adults, researchers can use common data collection tools such as questionnaires, interviews, and semi-structured interviews; yet, for young learners and very young ones, these reliable and valid data collection tools cannot be easily designed or applied by the researchers. In this study, firstly, common data collection tools are examined for ‘very young’ and ‘young learners’ participant groups since it is thought that the quality and efficiency of an academic study is mainly based on its valid and correct data collection and data analysis procedure. Secondly, two different data collection instruments for very young and young learners are stated as discussing the efficacy of them. Finally, a suggested data collection tool – a performance-based questionnaire- which is specifically developed for ‘very young’ and ‘young learners’ participant groups in the field of teaching English to young learners as a foreign language is presented in this current study. The designing procedure and suggested items/factors for the suggested data collection tool are accordingly revealed at the end of the study to help researchers have studied with young and very learners.

Keywords: data collection instruments, performance-based questionnaire, young learners, very young learners

Procedia PDF Downloads 92

Authors: Yaxin Bi

Abstract:

Accurate prediction and understanding of the evolution mechanisms of earthquakes remain challenging in the fields of geology, geophysics, and seismology. This study leverages Long Short-Term Memory (LSTM) networks and Generative Adversarial Networks (GANs), a generative model tailored to time-series data, for generating synthetic time series data based on Swarm satellite data, which will be used for detecting seismic anomalies. LSTMs demonstrated commendable predictive performance in generating synthetic data across multiple countries. In contrast, the GAN models struggled to generate synthetic data, often producing non-informative values, although they were able to capture the data distribution of the time series. These findings highlight both the promise and challenges associated with applying deep learning techniques to generate synthetic data, underscoring the potential of deep learning in generating synthetic electromagnetic satellite data.

Keywords: LSTM, GAN, earthquake, synthetic data, generative AI, seismic precursors

Procedia PDF Downloads 32

Authors: M. Qorbani

Abstract:

Neuromuscular control of posture as understood through studies of responses to mechanical sudden acceleration automatically has been previously demonstrated in individuals with chronic ankle instability (CAI), but the presence of acute condition has not been previously explored specially in a sudden acceleration. The aim of this study was to determine neuromuscular control pattern in those with and without unilateral acute ankle sprains. Design: Case - control. Setting: University research laboratory. The sinker–card protocol with surface translation was be used as a sudden acceleration protocol with study of EMG upon 4 posture stabilizer muscles in two sides of the body in response to sudden acceleration in forward and backward directions. 20 young adult women in two groups (10 LAS; 23.9 ± 2.03 yrs and 10 normal; 26.4 ± 3.2 yrs). The data of EMG were assessed by using multivariate test and one-way repeated measures 2×2×4 ANOVA (P< 0.05). The results showed a significant muscle by direction interaction. Higher TA activity of left and right side in LAS group than normal group in forward direction significantly be showed. Higher MGR activity in normal group than LAS group in backward direction significantly showed. These findings suggest that compared two sides of the body in two directions for 4 muscles EMG activities between and within group for neuromuscular control of posture in avoiding fall. EMG activations of two sides of the body in lateral ankle sprain (LAS) patients were symmetric significantly. Acute ankle instability following once ankle sprains caused to coordinated temporal spatial patterns and strategy selection.

Keywords: neuromuscular response, sEMG, lateral ankle sprain, posture.

Procedia PDF Downloads 482

Authors: Rory Brown, Jessica Daniels, Babatunde Oremule, William Tsang, Sadie Khwaja

Abstract:

Introduction: Dentures are common and are an intervention for both physical and psychological symptoms associated with tooth loss. However, the humble denture can cause morbidity and mortality if swallowed or aspirated. Numerous case reports document complications including hollow viscus perforation, fistula formation and airway compromise. The purpose of this review was to examine the literature documenting cases of swallowed or aspirated dentures over the past ten years to investigate factors that contribute to developing complications. Methods: A Medline literature search was performed to identify cases of denture ingestion or aspiration for over ten years. Data was collected to include patient, appliance and temporal factors that may contribute to developing complications including hollow viscus perforation, fistula formation, abscess, bowel obstruction, necrosis, hemorrhage and airway obstruction. The data was analyzed using observational and inferential statistics in the form of Chi-Squared and Pearson correlation tests. Results: Eighty-five cases of ingested or aspirated dentures were identified from 77 articles published between 1/10/2009 and 31/10/2019. Fourteen articles were excluded because they did not provide sufficient information on individual cases. Complications were documented in 37.6% of patients, and 2 cases resulted in death. There was no significant difference in complication risk based on patient age, hooked appliance, level of impaction, or radiolucency. However, symptoms of greater than 1-day duration are associated with an increased risk of complication (p=0.005). Increased time from ingestion or aspiration to removal is associated with an increased risk of complications, and the p-value remains significant up to and including day 4 (p=0.017). Conclusions: With denture use predicted to rise complications from the denture, ingestion and aspiration may become more frequent. We have demonstrated that increased symptom duration significantly increases the risk of developing complications. Additionally, we established the risk of developing complications is significantly reduced if the denture is removed with four days of aspiration or ingestion. By actively intervening early when presented with a case of swallowed or aspirated dentures, we may be able to reduce the morbidity associated with this unassuming device.

Keywords: aspiration, denture, ingestion, endoscopic foreign, body removal, foreign body impaction

Procedia PDF Downloads 138

Authors: Yogendra K. Karna, Lauren T. Bennett

Abstract:

Fire-tolerant eucalypt forests are one of the major forest ecosystems of south-eastern Australia and thought to be highly resistant to frequent high severity wildfires. However, the impact of different severity wildfires on the canopy structure of fire-tolerant forest type is under-studied, and there are significant knowledge gaps in relation to the assessment of tree and stand level canopy structural dynamics and recovery after fire. Assessment of canopy structure is a complex task involving accurate measurements of the horizontal and vertical arrangement of the canopy in space and time. This study examined the utility of multitemporal, small-footprint lidar data to describe the changes in the horizontal and vertical canopy structure of fire-tolerant eucalypt forests seven years after wildfire of different severities from the tree to stand level. Extensive ground measurements were carried out in four severity classes to describe and validate canopy cover and height metrics as they change after wildfire. Several metrics such as crown height and width, crown base height and clumpiness of crown were assessed at tree and stand level using several individual tree top detection and measurement algorithm. Persistent effects of high severity fire 8 years after both on tree crowns and stand canopy were observed. High severity fire increased the crown depth but decreased the crown projective cover leading to more open canopy.

Keywords: canopy gaps, canopy structure, crown architecture, crown projective cover, multi-temporal lidar, wildfire severity

Procedia PDF Downloads 175

Authors: K. S. Sreejith, C. Shaji

Abstract:

Though both Arabian Sea and its counterpart Bay of Bengal is forced primarily by the semi-annually reversing monsoons, the spatio-temporal variations of surface waters is very strong in the Arabian Sea as compared to the Bay of Bengal. This study focuses on the inter-annual variability of Sea Surface Temperature (SST) in the Arabian Sea by analysing ERSST dataset which covers 152 years of SST (January 1854 to December 2002) based on the ICOADS in situ observations. To capture the dominant SST oscillations and to understand the inter-annual SST variations at various local regions of the Arabian Sea, wavelet analysis was performed on this long time-series SST dataset. This tool is advantageous over other signal analysing tools like Fourier analysis, based on the fact that it unfolds a time-series data (signal) both in frequency and time domain. This technique makes it easier to determine dominant modes of variability and explain how those modes vary in time. The analysis revealed that pentadal SST oscillations predominate at most of the analysed local regions in the Arabian Sea. From the time information of wavelet analysis, it was interpreted that these cold and warm events of large amplitude occurred during the periods 1870-1890, 1890-1910, 1930-1950, 1980-1990 and 1990-2005. SST oscillations with peaks having period of ~ 2-4 years was found to be significant in the central and eastern regions of Arabian Sea. This indicates that the inter-annual SST variation in the Indian Ocean is affected by the El Niño-Southern Oscillation (ENSO) and Indian Ocean Dipole (IOD) events.

Keywords: Arabian Sea, ICOADS, inter-annual variation, pentadal oscillation, SST, wavelet analysis

Procedia PDF Downloads 276

Authors: Hyun-Woo Cho

Abstract:

For ensuring the safety of a manufacturing process one should quickly identify an assignable cause of a fault in an on-line basis. To this end, many statistical techniques including linear and nonlinear methods have been frequently utilized. However, such methods possessed a major problem of small sample size, which is mostly attributed to the characteristics of empirical models used for reference models. This work presents a new method to overcome the insufficiency of measurement data in the monitoring and diagnosis tasks. Some quasi-measurement data are generated from existing data based on the two indices of similarity and importance. The performance of the method is demonstrated using a real data set. The results turn out that the presented methods are able to handle the insufficiency problem successfully. In addition, it is shown to be quite efficient in terms of computational speed and memory usage, and thus on-line implementation of the method is straightforward for monitoring and diagnosis purposes.

Keywords: data analysis, diagnosis, monitoring, process data, quality control

Procedia PDF Downloads 481

Authors: Salehi Zahabi Saleh, Rajabi Hosaien, Rasaneh Samira

Abstract:

Yttrium-90 (90Y) radioembolisation(RE) is increasingly used for the treatment of patients with unresectable primary or metastatic liver tumours. Image-based approaches to assess microsphere distribution after RE have gained interest but are mostly hampered by the limited imaging possibilities of the Isotope 90Y. Quantitative 90Y-SPECT imaging has limited spatial resolution because it is based on 90Y Bremsstrahlung whereas 90Y-PET has better spatial resolution but low sensitivity. As a consequence, new alternative methods of visualizing the microspheres have been investigated, such as MR imaging of iron-labelled microspheres. It was also shown that MRI combines high sensitivity with high spatial and temporal resolution and with superior soft tissue contrast and thus can be used to cover a broad range of clinically interesting imaging parameters.The aim of the study in this article was to investigate the capability of MRI to measure the intrahepatic microsphere distribution in order to quantify the absorbed radiation dose in RE.

Keywords: radioembolisation, MRI, imaging, dosimetry

Procedia PDF Downloads 320

Authors: B. R. Campomanes-Alvarez, P. Quiros, B. Fernandez

Abstract:

Automatic Speech Recognition (ASR) is a machine-based process of decoding and transcribing oral speech. A typical ASR system receives acoustic input from a speaker or an audio file, analyzes it using algorithms, and produces an output in the form of a text. Some speech recognition systems use Hidden Markov Models (HMMs) to deal with the temporal variability of speech and Gaussian Mixture Models (GMMs) to determine how well each state of each HMM fits a short window of frames of coefficients that represents the acoustic input. Another way to evaluate the fit is to use a feed-forward neural network that takes several frames of coefficients as input and produces posterior probabilities over HMM states as output. Deep neural networks (DNNs) that have many hidden layers and are trained using new methods have been shown to outperform GMMs on a variety of speech recognition systems. Acoustic models for state-of-the-art ASR systems are usually training on massive amounts of data. However, audio files with their corresponding transcriptions can be difficult to obtain, especially in the Spanish language. Hence, in the case of these low-resource scenarios, building an ASR model is considered as a complex task due to the lack of labeled data, resulting in an under-trained system. Semi-supervised learning approaches arise as necessary tasks given the high cost of transcribing audio data. The main goal of this proposal is to develop a procedure based on acoustic semi-supervised learning for Spanish ASR systems by using DNNs. This semi-supervised learning approach consists of: (a) Training a seed ASR model with a DNN using a set of audios and their respective transcriptions. A DNN with a one-hidden-layer network was initialized; increasing the number of hidden layers in training, to a five. A refinement, which consisted of the weight matrix plus bias term and a Stochastic Gradient Descent (SGD) training were also performed. The objective function was the cross-entropy criterion. (b) Decoding/testing a set of unlabeled data with the obtained seed model. (c) Selecting a suitable subset of the validated data to retrain the seed model, thereby improving its performance on the target test set. To choose the most precise transcriptions, three confidence scores or metrics, regarding the lattice concept (based on the graph cost, the acoustic cost and a combination of both), was performed as selection technique. The performance of the ASR system will be calculated by means of the Word Error Rate (WER). The test dataset was renewed in order to extract the new transcriptions added to the training dataset. Some experiments were carried out in order to select the best ASR results. A comparison between a GMM-based model without retraining and the DNN proposed system was also made under the same conditions. Results showed that the semi-supervised ASR-model based on DNNs outperformed the GMM-model, in terms of WER, in all tested cases. The best result obtained an improvement of 6% relative WER. Hence, these promising results suggest that the proposed technique could be suitable for building ASR models in low-resource environments.

Keywords: automatic speech recognition, deep neural networks, machine learning, semi-supervised learning

Procedia PDF Downloads 339

Authors: Jiao F. Huang, Shi Chen, Shu C. Duan, Gang H. Wang

Abstract:

Due to its complex spatial structure as well as dynamic temporal evolution, an analytic solution of an X-pinch process is out of question, and numerical simulation becomes an important tool in X-pinch studies. Intrinsically, simulations of X-pinch are three-dimensional (3D) because of the specific structure of its load. Furthermore, in order to resolve both its μm-scales and ns-durations, fine spatial mesh grid and short time steps are usually adopted. The resulting large computational scales make the parallelization of codes a vital problem to be solved if any practical simulations are to be carried out. In this work, we report OpenMP parallelization of our 3D magnetohydrodynamic (MHD) code FOI-PERFECT. Results of test runs confirm that computational efficiency has been improved after parallelization, and both the sequential and parallel versions give the same physical results under the same initial conditions.

Keywords: MHD simulation, OpenMP, parallelization, X-pinch

Procedia PDF Downloads 340

Authors: Hsien-Tsen Wang

Abstract:

Business Intelligence (BI) has long helped organizations make informed decisions based on data-driven insights and gain competitive advantages in the marketplace. In the past two decades, businesses witnessed not only the dramatically increasing volume and heterogeneity of business data but also the emergence of new technologies, such as Artificial Intelligence (AI), Semantic Web (SW), Cloud Computing, and Big Data. It is plausible that the convergence of these technologies would bring more value out of business data by establishing linked data frameworks and connecting in ways that enable advanced analytics and improved data utilization. In this paper, we first review and summarize current BI applications and methodology. Emerging technologies that can be integrated into BI applications are then discussed. Finally, we conclude with a proposed synergy framework that aims at achieving a more flexible, scalable, and intelligent BI solution.

Keywords: business intelligence, artificial intelligence, semantic web, big data, cloud computing

Procedia PDF Downloads 94

Authors: John Q. Todd

Abstract:

Given that modern equipment can provide comprehensive health, status, and error condition data via built-in sensors, maintenance organizations have a new and valuable source of insight to take advantage of. This presentation will expose what these data payloads might look like and how they can be filtered, visualized, calculated into metrics, used for machine learning, and generate alerts for further action.

Keywords: condition based maintenance, equipment data, metrics, alerts

Procedia PDF Downloads 188

Authors: Ali Kazemi

Abstract:

The digital evolution of financial services, while offering unprecedented convenience and accessibility, has also escalated the vulnerabilities to fraudulent activities. In this study, we introduce a distinct approach to real-time anomaly detection in financial transactions, aiming to fortify the defenses of banking and financial institutions against such threats. Utilizing unsupervised machine learning algorithms, specifically autoencoders and isolation forests, our research focuses on identifying irregular patterns indicative of fraud within transactional data, thus enabling immediate action to prevent financial loss. The data we used in this study included the monetary value of each transaction. This is a crucial feature as fraudulent transactions may have distributions of different amounts than legitimate ones, such as timestamps indicating when transactions occurred. Analyzing transactions' temporal patterns can reveal anomalies (e.g., unusual activity in the middle of the night). Also, the sector or category of the merchant where the transaction occurred, such as retail, groceries, online services, etc. Specific categories may be more prone to fraud. Moreover, the type of payment used (e.g., credit, debit, online payment systems). Different payment methods have varying risk levels associated with fraud. This dataset, anonymized to ensure privacy, reflects a wide array of transactions typical of a global banking institution, ranging from small-scale retail purchases to large wire transfers, embodying the diverse nature of potentially fraudulent activities. By engineering features that capture the essence of transactions, including normalized amounts and encoded categorical variables, we tailor our data to enhance model sensitivity to anomalies. The autoencoder model leverages its reconstruction error mechanism to flag transactions that deviate significantly from the learned normal pattern, while the isolation forest identifies anomalies based on their susceptibility to isolation from the dataset's majority. Our experimental results, validated through techniques such as k-fold cross-validation, are evaluated using precision, recall, and the F1 score alongside the area under the receiver operating characteristic (ROC) curve. Our models achieved an F1 score of 0.85 and a ROC AUC of 0.93, indicating high accuracy in detecting fraudulent transactions without excessive false positives. This study contributes to the academic discourse on financial fraud detection and provides a practical framework for banking institutions seeking to implement real-time anomaly detection systems. By demonstrating the effectiveness of unsupervised learning techniques in a real-world context, our research offers a pathway to significantly reduce the incidence of financial fraud, thereby enhancing the security and trustworthiness of digital financial services.

Keywords: anomaly detection, financial fraud, machine learning, autoencoders, isolation forest, transactional data analysis

Procedia PDF Downloads 57

Authors: Dragana Calic

Abstract:

The openness, availability and the sheer volume of big data have provided, what some regard as, an invaluable and rich dataset. Researchers, businesses, advertising agencies, medical institutions, to name only a few, collect, share, and analyze this data to enable their processes and decision making. However, there are important ethical considerations associated with the use of big data. The rapidly evolving nature of online technologies has overtaken the many legislative, privacy, and ethical frameworks and principles that exist. For example, should we obtain consent to use people’s online data, and under what circumstances can privacy considerations be overridden? Current guidance on how to appropriately and ethically handle big data is inconsistent. Consequently, this paper focuses on two quite distinct but related ethical considerations that are at the core of the use of big data for research purposes. They include empowering the producers of data and empowering researchers who want to study big data. The first consideration focuses on informed consent which is at the core of empowering producers of data. In this paper, we discuss some of the complexities associated with informed consent and consider studies of producers’ perceptions to inform research ethics guidelines and practice. The second consideration focuses on the researcher. Similarly, we explore studies that focus on researchers’ perceptions and experiences.

Keywords: big data, ethics, producers’ perceptions, researchers’ perceptions

Procedia PDF Downloads 284