Search results for: attribute detection

Authors: Yihao Kuang, Bowen Ding

Abstract:

With the increasing scale and complexity of knowledge graph, modern knowledge graph contains more and more types of entity, relationship, and attribute information. Therefore, in recent years, it has been a trend for knowledge graph inference to use reinforcement learning to deal with large-scale, incomplete, and noisy knowledge graphs and improve the inference effect and interpretability. The Proximal Policy Optimization (PPO) algorithm utilizes a near-end strategy optimization approach. This allows for more extensive updates of policy parameters while constraining the update extent to maintain training stability. This characteristic enables PPOs to converge to improved strategies more rapidly, often demonstrating enhanced performance early in the training process. Furthermore, PPO has the advantage of offline learning, effectively utilizing historical experience data for training and enhancing sample utilization. This means that even with limited resources, PPOs can efficiently train for reinforcement learning tasks. Based on these characteristics, this paper aims to obtain a better and more efficient inference effect by introducing PPO into knowledge inference technology.

Keywords: reinforcement learning, PPO, knowledge inference

Procedia PDF Downloads 235

Authors: Jiri F. Urbanek, Jiri Barta, Oldrich Svoboda, Jiri J. Urbanek

Abstract:

The organizations of European and Czech critical infrastructure have specific position, mission, characteristics and behaviour in European Union and Czech state/ business environments, regarding specific requirements for regional and global security environments. They must respect policy of national security and global rules, requirements and standards in all their inherent and outer processes of supply-customer chains and networks. A controlling is generalized capability to have control over situational policy. This paper aims and purposes are to introduce the controlling as quite new necessary process attribute providing for critical infrastructure is environment the capability and profit to achieve its commitment regarding to the effectiveness of the quality management system in meeting customer/ user requirements and also the continual improvement of critical infrastructure organization’s processes overall performance and efficiency, as well as its societal security via continual planning improvement via DYVELOP modelling.

Keywords: added value, DYVELOP, controlling, environments, process approach

Procedia PDF Downloads 410

Authors: Hafid Smaili, Wilfred Rouwhorst, Paul Frost

Abstract:

The trend of recent accident and incident cases worldwide show that the state-of-the-art automation and operations, for current and future demanding operational environments, does not provide the desired level of operational safety under crew peak workload conditions, specifically in complex situations such as loss-of-control in-flight (LOC-I). Today, the short term focus is on preparing crews to recognise and handle LOC-I situations through upset recovery training. This paper describes the cockpit integration aspects and piloted assessment of both a manually assisted and automatic upset detection and recovery system that has been developed and demonstrated within the European Advanced Cockpit for Reduction Of StreSs and workload (ACROSS) programme. The proposed system is a function that continuously monitors and intervenes when the aircraft enters an upset and provides either manually pilot-assisted guidance or takes over full control of the aircraft to recover from an upset. In order to mitigate the highly physical and psychological impact during aircraft upset events, the system provides new cockpit functionalities to support the pilot in recovering from any upset both manually assisted and automatically. A piloted simulator assessment was made in Oct-Nov 2015 using ten pilots in a representative civil large transport fly-by-wire aircraft in terms of the preference of the tested upset detection and recovery system configurations to reduce pilot workload, increase situational awareness and safe interaction with the manually assisted or automated modes. The piloted simulator evaluation of the upset detection and recovery system showed that the functionalities of the system are able to support pilots during an upset. The experiment showed that pilots are willing to rely on the guidance provided by the system during an upset. Thereby, it is important for pilots to see and understand what the aircraft is doing and trying to do especially in automatic modes. Comparing the manually assisted and the automatic recovery modes, the pilot’s opinion was that an automatic recovery reduces the workload so that they could perform a proper screening of the primary flight display. The results further show that the manually assisted recoveries, with recovery guidance cues on the cockpit primary flight display, reduced workload for severe upsets compared to today’s situation. The level of situation awareness was improved for automatic upset recoveries where the pilot could monitor what the system was trying to accomplish compared to automatic recovery modes without any guidance. An improvement in situation awareness was also noticeable with the manually assisted upset recovery functionalities as compared to the current non-assisted recovery procedures. This study shows that automatic upset detection and recovery functionalities are likely to positively impact the operational safety by means of reduced workload, improved situation awareness and crew stress reduction. It is thus believed that future developments for upset recovery guidance and loss-of-control prevention should focus on automatic recovery solutions.

Keywords: aircraft accidents, automatic flight control, loss-of-control, upset recovery

Procedia PDF Downloads 206

Authors: Niddal Imam, Vassilios G. Vassilakis

Abstract:

After analysing messages posted by health-related spam campaigns in Twitter Arabic hashtags, we found that these campaigns use unique hijacked accounts (we call them adversarial hijacked accounts) as adversarial examples to fool deployed ML-based spam detectors. Existing ML-based models build a behaviour profile for each user to detect hijacked accounts. This approach is not applicable for detecting spam in Twitter hashtags since they are computationally expensive. Hence, we propose an adversary-aware ML-based detector, which includes a newly designed feature (avg posts) to improve the detection of spam tweets posted by the adversarial hijacked accounts at a tweet-level in trending hashtags. The proposed detector was designed considering three key points: robustness, adaptability, and interpretability. The new feature leverages the account’s temporal patterns (i.e., account age and number of posts). It is faster to compute compared to features discussed in the literature and improves the accuracy of detecting the identified hijacked accounts by 73%.

Keywords: Twitter spam detection, adversarial examples, evasion attack, adversarial concept drift, account hijacking, trending hashtag

Procedia PDF Downloads 72

Authors: Piyush Sharma, Devi Prasad Sharma, Sulabh Bansal

Abstract:

Diseases are fairly common in fruits and vegetables because of the changing climatic and environmental circumstances. Crop diseases, which are frequently difficult to control, interfere with the growth and output of the crops. Accurate disease detection and timely disease control measures are required to guarantee high production standards and good quality. In India, apples are a common crop that may be afflicted by a variety of diseases on the fruit, stem, and leaves. It is fungi, bacteria, and viruses that trigger the early symptoms of leaf diseases. In order to assist farmers and take the appropriate action, it is important to develop an automated system that can be used to detect the type of illnesses. Machine learning-based image processing can be used to: this research suggested a system that can automatically identify diseases in apple fruit and apple plants. Hence, this research utilizes the hybrid SVM-RBN model. As a consequence, the model may produce results that are more effective in terms of accuracy, precision, recall, and F1 Score, with respective values of 96%, 99%, 94%, and 93%.

Keywords: fruit plant disease, crop disease, machine learning, image processing, SVM-RBN

Procedia PDF Downloads 59

Authors: Dmitry Dikarev, Ajit Nimbalker, Alexei Davydov

Abstract:

Polar coding is a novel example of error correcting codes, which can achieve Shannon limit at block length N→∞ with log-linear complexity. Active research is being carried to adopt this theoretical concept for using in practical applications such as 5th generation wireless communication systems. Cyclic redundancy check (CRC) error detection code is broadly used in conjunction with successive cancellation list (SCL) decoding algorithm to improve finite-length polar code performance. However, there are two issues: increase of code block payload overhead by CRC bits and decrease of CRC error-detection capability. This paper proposes a method to control CRC overhead and false alarm rate of polar decoding. As shown in the computer simulations results, the proposed method provides the ability to use any set of CRC polynomials with any list size while maintaining the desired level of false alarm rate. This level of flexibility allows using polar codes in 5G New Radio standard.

Keywords: 5G New Radio, channel coding, cyclic redundancy check, list decoding, polar codes

Procedia PDF Downloads 234

Authors: G. Medkour Ishak-Boushaki, A. Taibi, M. Allab

Abstract:

Quantitative environmental radioactivity measurements are needed to determine the level of exposure of a population to ionizing radiations and for the assessment of the associated risks. Gamma spectrometry remains a very powerful tool for the analysis of radionuclides present in an environmental sample but the basic problem in such measurements is the low rate of detected events. Using large environmental samples could help to get around this difficulty but, unfortunately, new issues are raised by gamma rays attenuation and self-absorption. Recently, a new method has been suggested, to detect and identify without quantification, in a short time, a gamma ray of a low count source. This method does not require, as usually adopted in gamma spectrometry measurements, a pulse height spectrum acquisition. It is based on a chronological record of each detected photon by simultaneous measurements of its energy ε and its arrival time τ on the detector, the pair parameters [ε,τ] defining an event mode sequence (EMS). The EMS serials are analyzed sequentially by a Bayesian approach to detect the presence of a given radioactive source. The main object of the present work is to test the applicability of this sequential approach in radioactive environmental materials detection. Moreover, for an appropriate health oversight of the public and of the concerned workers, the analysis has been extended to get a reliable quantification of the radionuclides present in environmental samples. For illustration, we consider as an example, the problem of detection and quantification of 238U. Monte Carlo simulated experience is carried out consisting in the detection, by a Ge(Hp) semiconductor junction, of gamma rays of 63 keV emitted by 234Th (progeny of 238U). The generated EMS serials are analyzed by a Bayesian inference. The application of the sequential Bayesian approach, in environmental radioactivity analysis, offers the possibility of reducing the measurements time without requiring large environmental samples and consequently avoids the attached inconvenient. The work is still in progress.

Keywords: Bayesian approach, event mode sequence, gamma spectrometry, Monte Carlo method

Procedia PDF Downloads 493

Authors: El Harraj Abdeslam, Raissouni Naoufal

Abstract:

The detection of moving objects from a video image sequences is very important for object tracking, activity recognition, and behavior understanding in video surveillance. The most used approach for moving objects detection / tracking is background subtraction algorithms. Many approaches have been suggested for background subtraction. But, these are illumination change sensitive and the solutions proposed to bypass this problem are time consuming. In this paper, we propose a robust yet computationally efficient background subtraction approach and, mainly, focus on the ability to detect moving objects on dynamic scenes, for possible applications in complex and restricted access areas monitoring, where moving and motionless persons must be reliably detected. It consists of three main phases, establishing illumination changes in variance, background/foreground modeling and morphological analysis for noise removing. We handle illumination changes using Contrast Limited Histogram Equalization (CLAHE), which limits the intensity of each pixel to user determined maximum. Thus, it mitigates the degradation due to scene illumination changes and improves the visibility of the video signal. Initially, the background and foreground images are extracted from the video sequence. Then, the background and foreground images are separately enhanced by applying CLAHE. In order to form multi-modal backgrounds we model each channel of a pixel as a mixture of K Gaussians (K=5) using Gaussian Mixture Model (GMM). Finally, we post process the resulting binary foreground mask using morphological erosion and dilation transformations to remove possible noise. For experimental test, we used a standard dataset to challenge the efficiency and accuracy of the proposed method on a diverse set of dynamic scenes.

Keywords: video surveillance, background subtraction, contrast limited histogram equalization, illumination invariance, object tracking, object detection, behavior understanding, dynamic scenes

Procedia PDF Downloads 254

Authors: Rizka Ardhiyana, Liesbetini Haditjaroko, Sri Mulijani, Reki Ashadi Wicaksono, Raafqi Ranasasmita

Abstract:

Designing a sensitive, specific and easy to use method to detect pork contamination in the food industry remains a major challenge. In the current study, we developed a sensitive thiol-bond AuNP-Probe biosensor that will change color when detecting pork DNA in the Cytochrome B region. The interaction between the biosensors and DNA sample is measured by spectrophotometer at 540 nm. The biosensor is made by reducing gold with trisodium citrate to produce gold nanoparticle with 39.05 nm diameter. The AuNP-Probe biosensor (gold nanoprobe) achieved 16.04 ng DNA/µl limit of detection and 53.48 ng DNA/µl limit of quantification. The linearity (R2) between color absorbance changes and DNA concentration is 0.9916. The biosensor has a good specificty as it does not cross-react with DNA of chicken and beef. To verify specificity towards the target sequence, PCR was tested to the target sequence and reacted to the PCR product with the biosensor. The PCR DNA isolate resulted in a 2.7 fold higher absorbance compared to pork-DNA isolate alone (without PCR). The sensitivity and specificity of the method show the promising application of the thiol-bond AuNP biosensor in pork-detection.

Keywords: biosensor, DNA probe, gold nanoparticle (AuNP), pork meat, qPCR

Procedia PDF Downloads 356

Authors: Nilamadhab Mohanty

Abstract:

It is essential to use message formats that make communication understandable and correct. It is because; the information format can influence consumer decision on the purchase of a product. This study combines information from qualitative inquiry, media trend analysis, eye tracking experiment, and questionnaire data to examine the impact of specific message format and consumer perceived risk on attention to the information and risk retention. We investigated the influence of message framing (goal framing, attribute framing, and mix framing) on consumer memory, study time, and decisional uncertainty while deciding on the purchase of drugs. Furthermore, we explored the impact of consumer perceived risk (associated with the use of the drug, i.e., RISK-AB and perceived risk associated with the non-use of the drug, i.e., RISK-EB) on message format preference. The study used eye-tracking methods to understand the differences in message processing. Findings of the study suggest that the message format influences information processing, and participants' risk perception impacts message format preference. Eye tracking can be used to understand the format differences and design effective advertisements.

Keywords: message framing, consumer perceived risk, advertising, eye tracking

Procedia PDF Downloads 118

Authors: Said Abdul Ahad Ahadi, Elyas Baray, Nitin Rakesh, Sudeep Varshney

Abstract:

Wi-Fi is a widely used internet source that is used to provide internet access in many areas such as Stores, Cafes, University campuses, Restaurants and so on. This technology brought more facilities in communication and networking. On the other hand, due to the transmission of data over the air, which makes the network vulnerable, so it becomes prone to various threats such as Evil Twin and etc. The Evil Twin is a kind of adversary which impersonates a legitimate access point (LAP) as it can happen by spoofing the name (SSID) and MAC address (BSSID) of a legitimate access point (LAP). And this attack can cause many threats such as MITM, Service Interruption, Access point service blocking. Various Evil Twin Attack Detection Techniques are proposed, but they require additional hardware, or they require protocol modification. In this paper, we proposed a new technique based on Access Point’s two fingerprints, Received Signal Strength Indicator (RSSI) and Hop Count, that is hard to copy by an adversary. And we implemented the technique in a system called “ETDetector,” which can detect and prevent the attack.

Keywords: evil twin, LAP, SSID, Wi-Fi security, signal variation, ETAD, kali linux, scapy, python

Procedia PDF Downloads 140

Authors: Dylan Elliott, Katya Pertsova

Abstract:

Preposition errors are some of the most common errors created by L2 speakers. In addition, improving error correction and detection methods remains an open issue in the realm of Natural Language Processing (NLP). This research investigates whether the bidirectional encoder representations from the transformers model (BERT) have the potential to correct preposition errors accurately enough to be useful in error correction software. This research finds that BERT performs strongly when the scope of its error correction is limited to preposition choice. The researchers used an open-source BERT model and over three hundred thousand edited sentences from Wikipedia, tagged for part of speech, where only a preposition edit had occurred. To test BERT’s ability to detect errors, a technique known as multi-level masking was used to generate suggestions based on sentence context for every prepositional environment in the test data. These suggestions were compared with the original errors in the data and their known corrections to evaluate BERT’s performance. The suggestions were further analyzed to determine if BERT more often agreed with the judgements of the Wikipedia editors. Both the untrained and fined-tuned models were compared. Finetuning led to a greater rate of error-detection which significantly improved recall, but lowered precision due to an increase in false positives or falsely flagged errors. However, in most cases, these false positives were not errors in preposition usage but merely cases where more than one preposition was possible. Furthermore, when BERT correctly identified an error, the model largely agreed with the Wikipedia editors, suggesting that BERT’s ability to detect misused prepositions is better than previously believed. To evaluate to what extent BERT’s false positives were grammatical suggestions, we plan to do a further crowd-sourcing study to test the grammaticality of BERT’s suggested sentence corrections against native speakers’ judgments.

Keywords: BERT, grammatical error correction, preposition error detection, prepositions

Procedia PDF Downloads 142

Authors: C. Gopi Jinimole, A. Harsha

Abstract:

For detecting strokes, Computed Tomography (CT) scan is preferred for imaging the abnormalities or infarction in the brain. Because of the problems in the window settings used to evaluate brain CT images, they are very poor in the early stage infarction detection. This paper presents an automatic estimation method for the window settings of the CT images for proper contrast of the hyper infarction present in the brain. In the proposed work the window width is estimated automatically for each slice and the window centre is changed to a new value of 31HU, which is the average of the HU values of the grey matter and white matter in the brain. The automatic window width estimation is based on the average of median of statistical central moments. Thus with the new suggested window centre and estimated window width, the hyper infarction or post-stroke regions in CT brain images are properly detected. The proposed approach assists the radiologists in CT evaluation for early quantitative signs of delayed stroke, which leads to severe hemorrhage in the future can be prevented by providing timely medication to the patients.

Keywords: computed tomography (CT), hyper infarction or post stroke region, Hounsefield Unit (HU), window centre (WC), window width (WW)

Procedia PDF Downloads 198

Authors: Usman Qamar, Younus Javed

Abstract:

Frequent pattern mining is the process of finding a pattern (a set of items, subsequences, substructures, etc.) that occurs frequently in a data set. It was proposed in the context of frequent itemsets and association rule mining. Frequent pattern mining is used to find inherent regularities in data. What products were often purchased together? Its applications include basket data analysis, cross-marketing, catalog design, sale campaign analysis, Web log (click stream) analysis, and DNA sequence analysis. However, one of the bottlenecks of frequent itemset mining is that as the data increase the amount of time and resources required to mining the data increases at an exponential rate. In this investigation a new algorithm is proposed which can be uses as a pre-processor for frequent itemset mining. FASTER (FeAture SelecTion using Entropy and Rough sets) is a hybrid pre-processor algorithm which utilizes entropy and rough-sets to carry out record reduction and feature (attribute) selection respectively. FASTER for frequent itemset mining can produce a speed up of 3.1 times when compared to original algorithm while maintaining an accuracy of 71%.

Keywords: rough-sets, classification, feature selection, entropy, outliers, frequent itemset mining

Procedia PDF Downloads 432

Authors: Gopi Kandaswamy, P. Balamuralidhar

Abstract:

Fully autonomous small Unmanned Aerial Vehicles (UAVs) are increasingly being used in many commercial applications. Although a lot of research has been done to develop safe, reliable and durable UAVs, accidents due to electronic and structural failures are not uncommon and pose a huge safety risk to the UAV operators and the public. Hence there is a strong need for an automated health monitoring system for UAVs with a view to minimizing mission failures thereby increasing safety. This paper describes our approach to monitoring the electronic and structural components in a small UAV without the need for additional sensors to do the monitoring. Our system monitors data from four sources; sensors, navigation algorithms, control inputs from the operator and flight controller outputs. It then does statistical analysis on the data and applies a rule based engine to detect failures. This information can then be fed back into the UAV and a decision to continue or abort the mission can be taken automatically by the UAV and independent of the operator. Our system has been verified using data obtained from real flights over the past year from UAVs of various sizes that have been designed and deployed by us for various applications.

Keywords: fault detection, health monitoring, unmanned aerial vehicles, vibration analysis

Procedia PDF Downloads 257

Authors: Yihao Kuang, Bowen Ding

Abstract:

With the increasing scale and complexity of knowledge graph, modern knowledge graph contains more and more types of entity, relationship, and attribute information. Therefore, in recent years, it has been a trend for knowledge graph inference to use reinforcement learning to deal with large-scale, incomplete, and noisy knowledge graph and improve the inference effect and interpretability. The Proximal Policy Optimization (PPO) algorithm utilizes a near-end strategy optimization approach. This allows for more extensive updates of policy parameters while constraining the update extent to maintain training stability. This characteristic enables PPOs to converge to improve strategies more rapidly, often demonstrating enhanced performance early in the training process. Furthermore, PPO has the advantage of offline learning, effectively utilizing historical experience data for training and enhancing sample utilization. This means that even with limited resources, PPOs can efficiently train for reinforcement learning tasks. Based on these characteristics, this paper aims to obtain better and more efficient inference effect by introducing PPO into knowledge inference technology.

Keywords: reinforcement learning, PPO, knowledge inference, supervised learning

Procedia PDF Downloads 63

Authors: K. Jahani, J. Razavi

Abstract:

Nowadays, to decrease the number of downtimes in the industries such as metal mining, petroleum and chemical industries, predictive maintenance is crucial. In order to have efficient predictive maintenance, knowing the performance of critical equipment of production line such as pumps and hydro-cyclones under variable operating parameters, selecting best indicators of this equipment health situations, best locations for instrumentation, and also measuring of these indicators are very important. In this paper, computer aided engineering (CAE) tools are implemented to study some important elements of copper process line, namely slurry pumps and cyclone to predict the performance of these components under different working conditions. These modeling and simulations can be used in predicting, for example, the damage tolerance of the main shaft of the slurry pump or wear rate and location of cyclone wall or pump case and impeller. Also, the simulations can suggest best-measuring parameters, measuring intervals, and their locations.

Keywords: computer aided engineering, predictive maintenance, fault detection, mining process line, slurry pump, hydrocyclone

Procedia PDF Downloads 399

Authors: Simiao Ren, En Wei

Abstract:

Drowning is a significant safety issue worldwide, and a robust computer vision-based alert system can easily prevent such tragedies in swimming pools. However, due to domain shift caused by the visual gap (potentially due to lighting, indoor scene change, pool floor color etc.) between the training swimming pool and the test swimming pool, the robustness of such algorithms has been questionable. The annotation cost for labeling each new swimming pool is too expensive for mass adoption of such a technique. To address this issue, we propose a domain-aware data augmentation pipeline based on Gaussian Poisson Generative Adversarial Network (GP-GAN). Combined with YOLOv8, we demonstrate that such a domain adaptation technique can significantly improve the model performance (from 0.24 mAP to 0.82 mAP) on new test scenes. As the augmentation method only require background imagery from the new domain (no annotation needed), we believe this is a promising, practical route for preventing swimming pool drowning.

Keywords: computer vision, deep learning, YOLOv8, detection, swimming pool, drowning, domain adaptation, generative adversarial network, GAN, GP-GAN

Procedia PDF Downloads 90

Authors: Mojtaba Valinataj

Abstract:

Reversible logic as a new favorable design domain can be used for various fields especially creating quantum computers because of its speed and intangible power consumption. However, its susceptibility to a variety of environmental effects may lead to yield the incorrect results. In this paper, because of the importance of multiplication operation in various computing systems, some novel reversible logic array multipliers are proposed with error detection capability by incorporating the parity-preserving gates. The new designs are presented for two main parts of array multipliers, partial product generation and multi-operand addition, by exploiting the new arrangements of existing gates, which results in two signed parity-preserving array multipliers. The experimental results reveal that the best proposed 4×4 multiplier in this paper reaches 12%, 24%, and 26% enhancements in the number of constant inputs, number of required gates, and quantum cost, respectively, compared to previous design. Moreover, the best proposed design is generalized for n×n multipliers with general formulations to estimate the main reversible logic criteria as the functions of the multiplier size.

Keywords: array multipliers, Baugh-Wooley method, error detection, parity-preserving gates, quantum computers, reversible logic

Procedia PDF Downloads 255

Authors: Alireza Karimi

Abstract:

Artificial Intelligence (AI) is reshaping various industries, and accounting is no exception. With the ability to process vast amounts of data quickly and accurately, AI is revolutionizing how financial professionals manage, analyze, and report financial information. In this article, we will explore the diverse applications of AI in accounting and its profound impact on the field. Automation of Repetitive Tasks: One of the most significant contributions of AI in accounting is automating repetitive tasks. AI-powered software can handle data entry, invoice processing, and reconciliation with minimal human intervention. This not only saves time but also reduces the risk of errors, leading to more accurate financial records. Pattern Recognition and Anomaly Detection: AI algorithms excel at pattern recognition. In accounting, this capability is leveraged to identify unusual patterns in financial data that might indicate fraud or errors. AI can swiftly detect discrepancies, enabling auditors and accountants to focus on resolving issues rather than hunting for them. Real-Time Financial Insights: AI-driven tools, using natural language processing and computer vision, can process documents faster than ever. This enables organizations to have real-time insights into their financial status, empowering decision-makers with up-to-date information for strategic planning. Fraud Detection and Prevention: AI is a powerful tool in the fight against financial fraud. It can analyze vast transaction datasets, flagging suspicious activities and reducing the likelihood of financial misconduct going unnoticed. This proactive approach safeguards a company's financial integrity. Enhanced Data Analysis and Forecasting: Machine learning, a subset of AI, is used for data analysis and forecasting. By examining historical financial data, AI models can provide forecasts and insights, aiding businesses in making informed financial decisions and optimizing their financial strategies. Artificial Intelligence is fundamentally transforming the accounting profession. From automating mundane tasks to enhancing data analysis and fraud detection, AI is making financial processes more efficient, accurate, and insightful. As AI continues to evolve, its role in accounting will only become more significant, offering accountants and finance professionals powerful tools to navigate the complexities of modern finance. Embracing AI in accounting is not just a trend; it's a necessity for staying competitive in the evolving financial landscape.

Keywords: artificial intelligence, accounting automation, financial analysis, fraud detection, machine learning in finance

Procedia PDF Downloads 58

Authors: Marta Skwarecka, Patrycja Bloch, Rafal Walkusz, Oliwia Urbanowicz, Grzegorz Zielinski, Sabina Zoledowska, Dawid Nidzworski

Abstract:

Upper respiratory tract infections are one of the most common reasons for visiting a general doctor. Streptococci are the most common bacterial etiological factors in these infections. There are many different types of Streptococci and infections vary in severity from mild throat infections to pneumonia. For example, S. pyogenes mainly contributes to acute pharyngitis, palatine tonsils and scarlet fever, whereas S. Streptococcus pneumoniae is responsible for several invasive diseases like sepsis, meningitis or pneumonia with high mortality and dangerous complications. There are only a few diagnostic tests designed for detection Streptococci from the infected throat of patients. However, they are mostly based on lateral flow techniques, and they are not used as a standard due to their low sensitivity. The diagnostic standard is to culture patients throat swab on semi selective media in order to multiply pure etiological agent of infection and subsequently to perform antibiogram, which takes several days from the patients visit in the clinic. Therefore, the aim of our studies is to develop and implement to the market a Point of Care device for the rapid identification of Streptococcus pyogenes and Streptococcus pneumoniae with simultaneous identification of antibiotic resistance genes. In the course of our research, we successfully selected genes for to-species identification of Streptococci and genes encoding antibiotic resistance proteins. We have developed a reaction to amplify these genes, which allows detecting the presence of S. pyogenes or S. pneumoniae followed by testing their resistance to erythromycin, chloramphenicol and tetracycline. What is more, the detection of β-lactamase-encoding genes that could protect Streptococci against antibiotics from the ampicillin group, which are widely used in the treatment of this type of infection is also developed. The test is carried out directly from the patients' swab, and the results are available after 20 to 30 minutes after sample subjection, which could be performed during the medical visit.

Keywords: antibiotic resistance, Streptococci, respiratory infections, diagnostic test

Procedia PDF Downloads 123

Authors: Varsha Gupta, Priya Datta, Gursimran Mohi, Jagdish Chander

Abstract:

Fluoroquinolones form the mainstay of therapy for the treatment of infections due to Salmonella enterica subsp. enterica. There is a complex interplay between several resistance mechanisms for quinolones and various fluoroquinolones discs, giving varying results, making detection and interpretation of fluoroquinolone resistance difficult. For detection of fluoroquinolone resistance in Salmonella ssp., we compared the use of pefloxacin and nalidixic acid discs as surrogate marker. Using MIC for ciprofloxacin as the gold standard, 43.5% of strains showed MIC as ≥1 μg/ml and were thus resistant to fluoroquinoloes. Based on the performance of nalidixic acid and pefloxacin discs as surrogate marker for ciprofloxacin resistance, both the discs could correctly detect all the resistant phenotypes; however, use of nalidixic acid disc showed false resistance in the majority of the sensitive phenotypes. We have also tested newer antimicrobial agents like cefixime, imipenem, tigecycline and azithromycin against Salmonella spp. Moreover, there was a comeback of susceptibility to older antimicrobials like ampicillin, chloramphenicol, and cotrimoxazole. We can also use cefixime, imipenem, tigecycline and azithromycin in the treatment of multidrug resistant S. typhi due to their high susceptibility.

Keywords: salmonella, pefloxacin, surrogate marker, chloramphenicol

Procedia PDF Downloads 981

Authors: Zahra Ramezanpanah, Joachim Carvallo, Aurelien Rodriguez

Abstract:

This paper aims to detect anomalies in streaming data using machine learning algorithms. In this regard, we designed two separate pipelines and evaluated the effectiveness of each separately. The first pipeline, based on supervised machine learning methods, consists of two phases. In the first phase, we trained several supervised models using the UNSW-NB15 data-set. We measured the efficiency of each using different performance metrics and selected the best model for the second phase. At the beginning of the second phase, we first, using Argus Server, sniffed a local area network. Several types of attacks were simulated and then sent the sniffed data to a running algorithm at short intervals. This algorithm can display the results of each packet of received data in real-time using the trained model. The second pipeline presented in this paper is based on unsupervised algorithms, in which a Temporal Graph Network (TGN) is used to monitor a local network. The TGN is trained to predict the probability of future states of the network based on its past behavior. Our contribution in this section is introducing an indicator to identify anomalies from these predicted probabilities.

Keywords: temporal graph network, anomaly detection, cyber security, IDS

Procedia PDF Downloads 100

Authors: Anjali Singh, Ashwani Raju, Vandana Devi, Mohmad Mohsin Atique, Satyendra Singh, Munendra Singh

Abstract:

India is the biggest consumer of mercury and, consequently, a major emitter too. The increasing mercury contamination in India’s water resources has gained widespread attention and, therefore, atmospheric deposition is of critical concern. However, little emphasis was placed on the role of precipitation in the aquatic mercury cycle of the Ganga Alluvial Plain which provides drinking water to nearly 7% of the world’s human population. A majority of the precipitation here occurs primarily in 10% duration of the year in the monsoon season. To evaluate the sources and transportation of mercury, water sample analysis has been conducted from two selected sites near Lucknow, which have a strong hydraulic gradient towards the river. 31 groundwater samples from Jehta village (26°55’15’’N; 80°50’21’’E; 119 m above mean sea level) and 31 river water samples from the Behta Nadi (a tributary of the Gomati River draining into the Ganga River) were collected during the monsoon season on every alternate day between 01 July to 30 August 2019. The total mercury analysis was performed by using Flow Injection Atomic Absorption Spectroscopy (AAS)-Mercury Hybride System, and daily rainfall data was collected from the India Meteorological Department, Amausi, Lucknow. The ambient groundwater and river-water concentrations were both 2-4 ng/L as there is no known geogenic source of mercury found in the area. Before the onset of the monsoon season, the groundwater and the river-water recorded mercury concentrations two orders of magnitude higher than the ambient concentrations, indicating the regional transportation of the mercury from the non-point source into the aquatic environment. Maximum mercury concentrations in groundwater and river-water were three orders of magnitude higher than the ambient concentrations after the onset of the monsoon season characterizing the considerable mobilization and redistribution of mercury by monsoonal precipitation. About 50% of both of the water samples were reported mercury below the detection limit, which can be mostly linked to the low intensity of precipitation in August and also with the dilution factor by precipitation. The highest concentration ( > 1200 ng/L) of mercury in groundwater was reported after 6-days lag from the first precipitation peak. Two high concentration peaks (>1000 ng/L) in river-water were separately correlated with the surface flow and groundwater outflow of mercury. We attribute the elevated mercury concentration in both of the water samples before the precipitation event to mercury originating from the extensive use of agrochemicals in mango farming in the plain. However, the elevated mercury concentration during the onset of monsoon appears to increase in area wetted with atmospherically deposited mercury, which migrated down from surface water to groundwater as downslope migration is a fundamental mechanism seen in rivers of the alluvial plain. The present study underscores the significance of monsoonal precipitation in the transportation of mercury to drinking water resources of the Ganga Alluvial Plain. This study also suggests that future research must be pursued for a better understand of the human health impact of mercury contamination and for quantification of the role of Ganga Alluvial Plain in the Global Mercury Cycle.

Keywords: drinking water resources, Ganga alluvial plain, india, mercury

Procedia PDF Downloads 141

Authors: Gholba Niranjan Dilip, Anil Kumar

Abstract:

Conduct of accurate reliable mapping of oil palm plantations and census of individual palm trees is a huge challenge. This study addresses this challenge and developed an optimized solution implemented deep learning techniques on remote sensing data. The oil palm is a very important tropical crop. To improve its productivity and land management, it is imperative to have accurate census over large areas. Since, manual census is costly and prone to approximations, a methodology for automated census using panchromatic images from Cartosat-2, SkySat and World View-3 satellites is demonstrated. It is selected two different study sites in Indonesia. The customized set of training data and ground-truth data are created for this study from Cartosat-2 images. The pre-trained model of Single Shot MultiBox Detector (SSD) Lite MobileNet V2 Convolutional Neural Network (CNN) from the TensorFlow Object Detection API is subjected to transfer learning on this customized dataset. The SSD model is able to generate the bounding boxes for each oil palm and also do the counting of palms with good accuracy on the panchromatic images. The detection yielded an F-Score of 83.16 % on seven different images. The detections are buffered and dissolved to generate polygons demarcating the boundaries of the oil palm plantations. This provided the area under the plantations and also gave maps of their location, thereby completing the automated census, with a fairly high accuracy (≈100%). The trained CNN was found competent enough to detect oil palm crowns from images obtained from multiple satellite sensors and of varying temporal vintage. It helped to estimate the increase in oil palm plantations from 2014 to 2021 in the study area. The study proved that high-resolution panchromatic satellite image can successfully be used to undertake census of oil palm plantations using CNNs.

Keywords: object detection, oil palm tree census, panchromatic images, single shot multibox detector

Procedia PDF Downloads 157

Authors: Syed Beenish Rufai, Sarman Singh

Abstract:

Background: Performance of Genotype MTBDRsl (Hain Life science GmbH Germany) for detection of mutations associated with second-line drug resistance is well known. However, less evidence regarding the association of mutations and genetic background of strains is known which, in the future, is essential for clinical management of anti-tuberculosis drugs in those settings where the probability of particular genotype is predominant. Material and Methods: During this retrospective study, a total of 259 MDR-TB isolates obtained from pulmonary TB patients were tested for second-line drug susceptibility testing (DST) using Genotype MTBDRsl VER 1.0 and compared with BACTEC MGIT-960 as a reference standard. All isolates were further characterized using spoligotyping. The spoligo patterns obtained were compared and analyzed using SITVIT_WEB. Results: Of total 259 MDR-TB isolates which were screened for second-line DST by Genotype MTBDRsl, mutations were found to be associated with gyrA, rrs and emb genes in 82 (31.6%), 2 (0.8%) and 90 (34.7%) isolates respectively. 16 (6.1%) isolates detected mutations associated with both FQ as well as to AG/CP drugs (XDR-TB). No mutations were detected in 159 (61.4%) isolates for corresponding gyrA and rrs genes. Genotype MTBDRsl showed a concordance of 96.4% for detection of sensitive isolates in comparison with second-line DST by BACTEC MGIT-960 and 94.1%, 93.5%, 60.5% and 50% for detection of XDR-TB, FQ, EMB, and AMK/CAP respectively. D94G was the most prevalent mutation found among (38 (46.4%)) OFXR isolates (37 FQ mono-resistant and 1 XDR-TB) followed by A90V (23 (28.1%)) (17 FQ mono-resistant and 6 XDR-TB). Among AG/CP resistant isolates A1401G was the most frequent mutation observed among (11 (61.1%)) isolates (2 AG/CP mono-resistant isolates and 9 XDR-TB isolates) followed by WT+A1401G (6 (33.3%)) and G1484T (1 (5.5%)) respectively. On spoligotyping analysis, Beijing strain (46%) was found to be the most predominant strain among pre-XDR and XDR TB isolates followed by CAS (30%), X (6%), Unique (5%), EAI and T each of 4%, Manu (3%) and Ural (2%) respectively. Beijing strain was found to be strongly associated with D94G (47.3%) and A90V mutations by (47.3%) and 34.8% followed by CAS strain by (31.6%) and 30.4% respectively. However, among AG/CP resistant isolates, only Beijing strain was found to be strongly associated with A1401G and WT+A1401G mutations by 54.5% and 50% respectively. Conclusion: Beijing strain was found to be strongly associated with the most prevalent mutations among pre-XDR and XDR TB isolates. Acknowledgments: Study was supported with Grant by All India Institute of Medical Sciences, New Delhi reference No. P-2012/12452.

Keywords: tuberculosis, line probe assay, XDR TB, drug susceptibility

Procedia PDF Downloads 138

Authors: Manal Kamel, Sara Maher

Abstract:

Development of sensitive non-invasive tests for detection of SARS-CoV-2 antigens is imperative to manage the extent of infection throughout the population, yet, it is still challenging. Here, we designed and optimized a sandwich enzyme-linked immunosorbent assay (ELISA) for SARS-CoV-2 S1 antigen detection in saliva. Both saliva samples and nasopharyngeal swapswere collected from 170 PCR-confirmed positive and negative cases. Gold nanoparticles (AuNPs) were conjugated with S1protein receptor binding domain (RBD) nanobodies. Recombinant S1 monoclonal antibodies (S1mAb) as primery antibody and gold conjugated nanobodies as secondary antibody were employed in sandwich ELISA. Our developed system were optimized to achieve 87.5 % sensitivity and 100% specificity for saliva samples compared to 89 % and 100% for nasopharyngeal swaps, respectively. This means that saliva could be a suitable replacement for nasopharyngeal swaps No cross reaction was detected with other corona virus antigens. These results revealed that our developed ELISAcould be establishedas a new, reliable, sensitive, and non-invasive test for diagnosis of SARS-CoV-2 infection, using the easily collected saliva samples.

Keywords: COVID 19, diagnosis, ELISA, nanobodies

Procedia PDF Downloads 130

Authors: Shih-Yu Lo

Abstract:

Stereotypes exist in almost every society, affecting how people interact with each other. However, to our knowledge, the influence of stereotypes was rarely explored in the context of basic perceptual processes. This study aims to explore how the gender stereotype affects object recognition. Participants were presented with a series of scene pictures, followed by a target display with a man or a woman, holding a weapon or a non-weapon object. The task was to identify whether the object in the target display was a weapon or not. Although the gender of the object holder could not predict whether he or she held a weapon, and was irrelevant to the task goal, the participant nevertheless tended to identify the object as a weapon when the object holder was a man than a woman. The analysis based on the signal detection theory showed that the stereotype effect on object recognition mainly resulted from the participant’s bias to make a 'weapon' response when a man was in the scene instead of a woman in the scene. In addition, there was a trend that the participant’s sensitivity to differentiate a weapon from a non-threating object was higher when a woman was in the scene than a man was in the scene. The results of this study suggest that the irrelevant social cues implied in the visual scene can be very powerful that they can modulate the basic object recognition process.

Keywords: gender stereotype, object recognition, signal detection theory, weapon

Procedia PDF Downloads 203

Authors: Muhammad R. Ahmed, Mohammed Aseeri

Abstract:

Wireless Sensor Network (WSN) consists of low-cost and multi functional resources constrain nodes that communicate at short distances through wireless links. It is open media and underpinned by an application driven technology for information gathering and processing. It can be used for many different applications range from military implementation in the battlefield, environmental monitoring, health sector as well as emergency response of surveillance. With its nature and application scenario, security of WSN had drawn a great attention. It is known to be valuable to variety of attacks for the construction of nodes and distributed network infrastructure. In order to ensure its functionality especially in malicious environments, security mechanisms are essential. Malicious or internal attacker has gained prominence and poses the most challenging attacks to WSN. Many works have been done to secure WSN from internal attacks but most of it relay on either training data set or predefined threshold. Without a fixed security infrastructure a WSN needs to find the internal attacks is a challenge. In this paper we present an internal attack detection method based on maximum entropy model. The final experimental works showed that the proposed algorithm does work well at the designed level.

Keywords: internal attack, wireless sensor network, network security, entropy

Procedia PDF Downloads 452

Authors: Marcel Chevalier, Léo Dupont, Sylvain Marié, Frédérique Roffet, Elena Stolyarova, William Templier, Costin Vasile

Abstract:

Industrial end-users are currently facing an increasing need to reduce the risk of unexpected failures and optimize their maintenance. This calls for both short-term analysis and long-term ageing anticipation. At Schneider Electric, we tackle those two issues using both machine learning and first principles models. Machine learning models are incrementally trained from normal data to predict expected values and detect statistically significant short-term deviations. Ageing models are constructed by breaking down physical systems into sub-assemblies, then determining relevant degradation modes and associating each one to the right kinetic law. Validating such anomaly detection and maintenance models is challenging, both because actual incident and ageing data are rare and distorted by human interventions, and incremental learning depends on human feedback. To overcome these difficulties, we propose to simulate physics, systems, and humans -including asset maintenance operations- in order to validate the overall approaches in accelerated time and possibly choose between algorithmic alternatives.

Keywords: degradation models, ageing, anomaly detection, soft sensor, incremental learning

Procedia PDF Downloads 123