Search results for: classification

Authors: Nishtha Madaan, Prateek Chaudhury, Nishant Kumar, Srikanta Bedathur

Abstract:

In modern NLP applications, word embeddings are a crucial backbone that can be readily shared across a number of tasks. However, as the text distributions change and word semantics evolve over time, the downstream applications using the embeddings can suffer if the word representations do not conform to the data drift. Thus, maintaining word embeddings to be consistent with the underlying data distribution is a key problem. In this work, we tackle this problem and propose TransDrift, a transformer-based prediction model for word embeddings. Leveraging the flexibility of the transformer, our model accurately learns the dynamics of the embedding drift and predicts future embedding. In experiments, we compare with existing methods and show that our model makes significantly more accurate predictions of the word embedding than the baselines. Crucially, by applying the predicted embeddings as a backbone for downstream classification tasks, we show that our embeddings lead to superior performance compared to the previous methods.

Keywords: NLP applications, transformers, Word2vec, drift, word embeddings

Procedia PDF Downloads 91

Authors: Alberto Alcalá-Alvarez, Pablo Padilla-Longoria

Abstract:

The present paper describes a framework for constructing mathematical networks encoding relevant musical information from a music score for structural analysis. These graphs englobe statistical information about music elements such as notes, chords, rhythms, intervals, etc., and the relations among them, and so become helpful in visualizing and understanding important stylistic features of a music fragment. In order to build such networks, musical data is parsed out of a digital symbolic music file. This data undergoes different analytical procedures from Graph Theory, such as measuring the centrality of nodes, community detection, and entropy calculation. The resulting networks reflect important structural characteristics of the fragment in question: predominant elements, connectivity between them, and complexity of the information contained in it. Music pieces in different styles are analyzed, and the results are contrasted with the traditional analysis outcome in order to show the consistency and potential utility of this method for music analysis.

Keywords: computational musicology, mathematical music modelling, music analysis, style classification

Procedia PDF Downloads 102

Authors: A. Biran, P. Sobhe Bidari, A. Almazroe, V. Lakshminarayanan, K. Raahemifar

Abstract:

Diabetic Retinopathy (DR) is a severe retinal disease which is caused by diabetes mellitus. It leads to blindness when it progress to proliferative level. Early indications of DR are the appearance of microaneurysms, hemorrhages and hard exudates. In this paper, an automatic algorithm for detection of DR has been proposed. The algorithm is based on combination of several image processing techniques including Circular Hough Transform (CHT), Contrast Limited Adaptive Histogram Equalization (CLAHE), Gabor filter and thresholding. Also, Support Vector Machine (SVM) Classifier is used to classify retinal images to normal or abnormal cases including non-proliferative or proliferative DR. The proposed method has been tested on images selected from Structured Analysis of the Retinal (STARE) database using MATLAB code. The method is perfectly able to detect DR. The sensitivity specificity and accuracy of this approach are 90%, 87.5%, and 91.4% respectively.

Keywords: diabetic retinopathy, fundus images, STARE, Gabor filter, support vector machine

Procedia PDF Downloads 294

Authors: Deniz Peksen

Abstract:

This thesis aims to predict trend movement of closing price of stock and to maximize portfolio by utilizing the predictions. In this context, the study aims to define a stock portfolio strategy from models created by using Logistic Regression, Gradient Boosting and Random Forest. Recently, predicting the trend of stock price has gained a significance role in making buy and sell decisions and generating returns with investment strategies formed by machine learning basis decisions. There are plenty of studies in the literature on the prediction of stock prices in capital markets using machine learning methods but most of them focus on closing prices instead of the direction of price trend. Our study differs from literature in terms of target definition. Ours is a classification problem which is focusing on the market trend in next 20 trading days. To predict trend direction, fourteen years of data were used for training. Following three years were used for validation. Finally, last three years were used for testing. Training data are between 2002-06-18 and 2016-12-30 Validation data are between 2017-01-02 and 2019-12-31 Testing data are between 2020-01-02 and 2022-03-17 We determine Hold Stock Portfolio, Best Stock Portfolio and USD-TRY Exchange rate as benchmarks which we should outperform. We compared our machine learning basis portfolio return on test data with return of Hold Stock Portfolio, Best Stock Portfolio and USD-TRY Exchange rate. We assessed our model performance with the help of roc-auc score and lift charts. We use logistic regression, Gradient Boosting and Random Forest with grid search approach to fine-tune hyper-parameters. As a result of the empirical study, the existence of uptrend and downtrend of five stocks could not be predicted by the models. When we use these predictions to define buy and sell decisions in order to generate model-based-portfolio, model-based-portfolio fails in test dataset. It was found that Model-based buy and sell decisions generated a stock portfolio strategy whose returns can not outperform non-model portfolio strategies on test dataset. We found that any effort for predicting the trend which is formulated on stock price is a challenge. We found same results as Random Walk Theory claims which says that stock price or price changes are unpredictable. Our model iterations failed on test dataset. Although, we built up several good models on validation dataset, we failed on test dataset. We implemented Random Forest, Gradient Boosting and Logistic Regression. We discovered that complex models did not provide advantage or additional performance while comparing them with Logistic Regression. More complexity did not lead us to reach better performance. Using a complex model is not an answer to figure out the stock-related prediction problem. Our approach was to predict the trend instead of the price. This approach converted our problem into classification. However, this label approach does not lead us to solve the stock prediction problem and deny or refute the accuracy of the Random Walk Theory for the stock price.

Keywords: stock prediction, portfolio optimization, data science, machine learning

Procedia PDF Downloads 80

Authors: Mustafa M. Bob, Norhan Rahman, Abdalla Elamin, Saud Taher

Abstract:

The objective of this study was to assess the suitability of Madinah city groundwater for irrigation purposes. Of the twenty three wells that were drilled in different locations in the city for the purposes of this study, twenty wells were sampled for water quality analyses. The United States Department of Agriculture (USDA) classification of irrigation water that is based on Sodium hazard (SAR) and salinity hazard was used for suitability assessment. In addition, the residual sodium carbonate (RSC) was calculated for all samples and also used for irrigation suitability assessment. Results showed that all groundwater samples are in the acceptable quality range for irrigation based on RSC values. When SAR and salinity hazard were assessed, results showed that while all groundwater samples (except one) fell in the acceptable range of SAR, they were either in the high or very high salinity zone which indicates that care should be taken regarding the type of soil and crops in the study area.

Keywords: irrigation suitability, TDS, salinity, SAR

Procedia PDF Downloads 372

Authors: Monika Chuchro

Abstract:

This paper presents a preliminary attempt to apply classification of time series using meta-clusters in order to improve the quality of regression models. In this case, clustering was performed as a method to obtain a subgroups of time series data with normal distribution from inflow into waste water treatment plant data which Composed of several groups differing by mean value. Two simple algorithms: K-mean and EM were chosen as a clustering method. The rand index was used to measure the similarity. After simple meta-clustering, regression model was performed for each subgroups. The final model was a sum of subgroups models. The quality of obtained model was compared with the regression model made using the same explanatory variables but with no clustering of data. Results were compared by determination coefficient (R2), measure of prediction accuracy mean absolute percentage error (MAPE) and comparison on linear chart. Preliminary results allows to foresee the potential of the presented technique.

Keywords: clustering, data analysis, data mining, predictive models

Procedia PDF Downloads 466

Authors: Maha Wiss, Wael Khreich

Abstract:

Animals, objects, foods, plants, and other non-human terms are commonly used as a source of metaphors to describe females in formal and slang language. Comparing women to non-human items not only reflects cultural views that might conceptualize women as subordinates or in a lower position than humans, yet it conveys this degradation to the listeners. Moreover, the dehumanizing representation of females in the language normalizes the derogation and even encourages sexism and aggressiveness against women. Although dehumanization has been a popular research topic for decades, according to our knowledge, no studies have linked women's dehumanizing language to the machine learning field. Therefore, we introduce our research work as one of the first attempts to create a tool for the automated detection of the dehumanizing depiction of females in English texts. We also present the first labeled dataset on the charted topic, which is used for training supervised machine learning algorithms to build an accurate classification model. The importance of this work is that it accomplishes the first step toward mitigating dehumanizing language against females.

Keywords: gender bias, machine learning, NLP, women dehumanization

Procedia PDF Downloads 80

Authors: Ines Gasmi, Salima Smiti, Makram Soui, Khaled Ghedira

Abstract:

Credit risk is considered as one of the important issues for financial institutions. It provokes great losses for banks. To this objective, numerous methods for credit risk evaluation have been proposed. Many evaluation methods are black box models that cannot adequately reveal information hidden in the data. However, several works have focused on building transparent rules-based models. For credit risk assessment, generated rules must be not only highly accurate, but also highly interpretable. In this paper, we aim to build both, an accurate and transparent credit risk evaluation model which proposes a set of classification rules. In fact, we consider the credit risk evaluation as an optimization problem which uses a genetic programming (GP) algorithm, where the goal is to maximize the accuracy of generated rules. We evaluate our proposed approach on the base of German and Australian credit datasets. We compared our finding with some existing works; the result shows that the proposed GP outperforms the other models.

Keywords: credit risk assessment, rule generation, genetic programming, feature selection

Procedia PDF Downloads 353

Authors: Chuluundorj Begz

Abstract:

This paper presents the dynamic, psycho-cognitive approach to study of human verbal thinking on the basis of typologically different languages /as a Mongolian, English and Russian/. Topological equivalence in verbal communication serves as a basis of Universality of mental structures and therefore deep structures. Mechanism of verbal thinking consisted at the deep level of basic concepts, rules for integration and classification, neural networks of vocabulary. In neuro cognitive study of language, neural architecture and neuro psychological mechanism of verbal cognition are basis of a vector-based modeling. Verbal perception and interpretation of the infinite set of meanings and propositions in mental continuum can be modeled by applying tensor methods. Euclidean and non-Euclidean spaces are applied for a description of human semantic vocabulary and high order structures.

Keywords: Euclidean spaces, isomorphism and homomorphism, mental lexicon, mental mapping, semantic memory, verbal cognition, vector space

Procedia PDF Downloads 519

Authors: U.V. Suryawanshi, S.S. Chowhan, U.V Kulkarni

Abstract:

Accuracy of segmentation methods is of great importance in brain image analysis. Tissue classification in Magnetic Resonance brain images (MRI) is an important issue in the analysis of several brain dementias. This paper portraits performance of segmentation techniques that are used on Brain MRI. A large variety of algorithms for segmentation of Brain MRI has been developed. The objective of this paper is to perform a segmentation process on MR images of the human brain, using Fuzzy c-means (FCM), Kernel based Fuzzy c-means clustering (KFCM), Spatial Fuzzy c-means (SFCM) and Improved Fuzzy c-means (IFCM). The review covers imaging modalities, MRI and methods for noise reduction and segmentation approaches. All methods are applied on MRI brain images which are degraded by salt-pepper noise demonstrate that the IFCM algorithm performs more robust to noise than the standard FCM algorithm. We conclude with a discussion on the trend of future research in brain segmentation and changing norms in IFCM for better results.

Keywords: image segmentation, preprocessing, MRI, FCM, KFCM, SFCM, IFCM

Procedia PDF Downloads 331

Authors: Jeewan Rai, Kinzang, Yeshi Jigme Choden

Abstract:

Accumulation of dust on solar panels impacts the overall efficiency and the amount of energy they produce. While various techniques exist for detecting dust to schedule cleaning, many of these methods use MATLAB image processing tools and other licensed software, which can be financially burdensome. This study will investigate the efficiency of a free open-source computer vision library using the YOLO algorithm. The proposed approach has been tested on images of solar panels with varying dust levels through an experiment setup. The experimental findings illustrated the effectiveness of using the YOLO-based image classification method and the overall dust detection approach with an accuracy of 90% in distinguishing between clean and dusty panels. This open-source solution provides a cost effective and accessible alternative to commercial image processing tools, offering solutions for optimizing solar panel maintenance and enhancing energy production.

Keywords: YOLO, openCV, dust detection, solar panels, computer vision, image processing

Procedia PDF Downloads 32

Authors: Hamed Sarbazy, Zohre Gholipour Haftkhani, Ali Safari, Pejman Hosseiniun

Abstract:

Scientific and industrial progress in the past two decades has resulted in energy distribution systems based on power electronics, as an enabling technology in various industries and building management systems can be considered. Grading and standardization module power electronics systems and its use in a distributed control system, a strategy for overcoming the limitations of using this system. The purpose of this paper is to investigate strategies for scheduling and control structure of standard modules is a power electronic systems. This paper introduces the classical control methods and disadvantages of these methods will be discussed, The hierarchical control as a mechanism for distributed control structure of the classification module explains. The different levels of control and communication between these levels are fully introduced. Also continue to standardize software distribution system control structure is discussed. Finally, as an example, the control structure will be presented in a DC distribution system.

Keywords: application management, hardware management, power electronics, building blocks

Procedia PDF Downloads 521

Authors: Gemma García López

Abstract:

The goal of this study is to analyze if search queries carried out in search engines such as Google, can offer emotional information about the user that performs them. Knowing the emotional state in which the Internet user is located can be a key to achieve the maximum personalization of content and the detection of worrying behaviors. For this, two studies were carried out using tools with advanced natural language processing techniques. The first study determines if a query can be classified as positive, negative or neutral, while the second study extracts emotional content from words and applies the categorical and dimensional models for the representation of emotions. In addition, we use search queries in Spanish and English to establish similarities and differences between two languages. The results revealed that text search queries performed by users on the Internet can be classified emotionally. This allows us to better understand the emotional state of the user at the time of the search, which could involve adapting the technology and personalizing the responses to different emotional states.

Keywords: emotion classification, text search queries, emotional analysis, sentiment analysis in text, natural language processing

Procedia PDF Downloads 141

Authors: Hamed Hakim, Roderic Archambault, Charles J. Kibert, Maryam Mirhadi Fard

Abstract:

Changes in the patterns of life in the late 20th and early 21st century have created new challenges for educational systems. Greening the physical environment of school buildings has emerged as a response to some of those challenges and led to the design of energy efficient K-12 school buildings. With the advancement in knowledge and technology, the successful construction of Net Zero Energy Schools, such as the Lady Bird Johnson Middle School demonstrates a cutting edge generation of sustainable schools, and solves the former challenge of attaining energy self-sufficient educational facilities. There are approximately twenty net zero energy K-12 schools in the U.S. of which about six are located in Climate Zone 5 and 6 based on ASHRAE climate zone classification. This paper aims to describe and analyze the current status of energy efficient and NZE schools in Canada. An attempt is made to study existing U.S. energy neutral strategies closest to the climate zones in Canada (zones 5 and 6) and identify the best practices for Canadian schools.

Keywords: Canada K-12 schools, green school, energy efficient, net-zero energy schools

Procedia PDF Downloads 404

Authors: Edmund Gyimah

Abstract:

Using data from 146 annual reports of listed firms in Ghana for the period 2013-2020, this study presents indicative findings which inspire practical actions and future research. Firms which prepared and presented sustainability reports were excluded from this study for a coverage of corporate sustainability disclosures centred on annual reports. Also, corporate sustainability disclosures of the firms on corporate websites were not included in the study considering the tendency of updates which cannot easily be traced. The corporate sustainability disclosures in the annual reports since the commencement of the G4 Guidelines in 2013 have been below average for all the dimensions of sustainability and the general sustainability disclosures. Few traditional elements of the board composition such as board size and board independence could affect the corporate sustainability disclosures in the annual reports as well as the age of the firm, firm size, and industry classification of the firm. Sustainability disclosures are greater in sustainability reports than in annual reports, however, firms without sustainability reports should have a considerable amount of sustainability disclosures in their annual reports. Also, because of the essence of sustainability, this study suggests to firms to have sustainability committee perhaps, they could make a difference in disclosing the enough sustainability information even when they do not present sustainability information in stand-alone reports.

Keywords: disclosures, sustainability, board, reports

Procedia PDF Downloads 188

Authors: Milad Vahidi, Mahmod R. Sahebi, Mehrnoosh Omati, Reza Mohammadi

Abstract:

Forest management organizations need information to perform their work effectively. Remote sensing is an effective method to acquire information from the Earth. Two datasets of remote sensing images were used to classify forested regions. Firstly, all of extractable features from hyperspectral and PolSAR images were extracted. The optical features were spectral indexes related to the chemical, water contents, structural indexes, effective bands and absorption features. Also, PolSAR features were the original data, target decomposition components, and SAR discriminators features. Secondly, the particle swarm optimization (PSO) and the genetic algorithms (GA) were applied to select optimization features. Furthermore, the support vector machine (SVM) classifier was used to classify the image. The results showed that the combination of PSO and SVM had higher overall accuracy than the other cases. This combination provided overall accuracy about 90.56%. The effective features were the spectral index, the bands in shortwave infrared (SWIR) and the visible ranges and certain PolSAR features.

Keywords: hyperspectral, PolSAR, feature selection, SVM

Procedia PDF Downloads 416

Authors: Jing Zhang

Abstract:

To reduce the carbon footprint of mankind and become more sustainable is one of the major challenges in our era. Internet of Things (IoT) mainly resolves three problems: Things to Things (T2T), Human to Things, H2T), and Human to Human (H2H). Borrowing the classification of IoT, we can find carbon prints of industries also can be divided in these three ways. Therefore, monitoring the routes of generation and circulation of products may help calculate product carbon print. This paper does not consider any technique used by IoT itself, but the ideas of it look at the connection of products. Carbon prints are like a gene or mark of a product from raw materials to the final products, which never leave the products. The contribution of this paper is to combine the characteristics of IoT and the methodology of network science to find a way to calculate the product's carbon footprint. Life cycle assessment, LCA is a traditional and main tool to calculate the carbon print of products. LCA is a traditional but main tool, which includes three kinds.

Keywords: product carbon footprint, Internet of Things, network science, life cycle assessment

Procedia PDF Downloads 116

Authors: Sanaz Kavianpour, Zuraini Ismail, Bharanidharan Shanmugam

Abstract:

Social Network Sites (SNSs) can be served as an invaluable platform to transfer the information across a large number of individuals. A substantial component of communicating and managing information is to identify which individual will influence others in propagating information and also whether dissemination of information in the absence of social signals about that information will be occurred or not. Classifying the final audience of social data is difficult as controlling the social contexts which transfers among individuals are not completely possible. Hence, undesirable information diffusion to an unauthorized individual on SNSs can threaten individuals’ privacy. This paper highlights the information diffusion in SNSs and moreover it emphasizes the most significant privacy issues to individuals of SNSs. The goal of this paper is to propose a privacy-preserving model that has urgent regards with individuals’ data in order to control availability of data and improve privacy by providing access to the data for an appropriate third parties without compromising the advantages of information sharing through SNSs.

Keywords: anonymization algorithm, classification algorithm, information diffusion, privacy, social network sites

Procedia PDF Downloads 321

Authors: Radoslav Židek, Veronika Šidlová, Radovan Kasarda, Birgit Fuerst-Waltl

Abstract:

Machine learning represents a set of topics dealing with the creation and evaluation of algorithms that facilitate pattern recognition, classification, and prediction, based on models derived from existing data. The data can present identification patterns which are used to classify into groups. The result of the analysis is the pattern which can be used for identification of data set without the need to obtain input data used for creation of this pattern. An important requirement in this process is careful data preparation validation of model used and its suitable interpretation. For breeders, it is important to know the origin of animals from the point of the genetic diversity. In case of missing pedigree information, other methods can be used for traceability of animal´s origin. Genetic diversity written in genetic data is holding relatively useful information to identify animals originated from individual countries. We can conclude that the application of data mining for molecular genetic data using supervised learning is an appropriate tool for hypothesis testing and identifying an individual.

Keywords: genetic data, Pinzgau cattle, supervised learning, machine learning

Procedia PDF Downloads 550

Authors: Moulay Smaïne Ghembaza, Mokhtar Dadouch, Nour-Said Ikhlef

Abstract:

Soil treatment is to make use soil that does not have the characteristics required in a given context. We limit ourselves in this work to the field of road earthworks where we have chosen to develop a local material in the region of Sidi Bel Abbes (Algeria). This material has poor characteristics not meeting the standards used in road geo technics. To remedy this, firstly, we were trying to improve the Proctor Standard characteristics of this material by mechanical treatment increasing the compaction energy. Then, by a chemical treatment, adding some cement dosages, our results show that this material classified A1h a increase maximum dry density and a reduction in the water content of compaction. A comparative study is made on the optimal properties of the material between the two modes of treatment. On the other hand, after treatment, one finds a decrease in the plasticity index and the methylene blue value. This material exhibits a change of class. Therefore, soil class CL turned into a soil class composed CL-ML (Silt of low plasticity). This observation allows this material to be used as backfill or sub grade.

Keywords: treatment of soil, cement, subgrade, Atteberg limits, classification, optimum proctor properties

Procedia PDF Downloads 472

Authors: Vaishnavi Shivde, Shrishti Sinha, Trapti Mishra

Abstract:

Traffic light detection from a moving vehicle is an important technology both for driver safety assistance functions as well as for autonomous driving in the city. This paper proposed a deep-learning-based traffic light recognition method that consists of a pixel-wise image segmentation technique and a fully convolutional network i.e., UNET architecture. This paper has used a method for detecting the position and recognizing the state of the traffic lights in video sequences is presented and evaluated using Traffic Light Dataset which contains masked traffic light image data. The first stage is the detection, which is accomplished through image processing (image segmentation) techniques such as image cropping, color transformation, segmentation of possible traffic lights. The second stage is the recognition, which means identifying the color of the traffic light or knowing the state of traffic light which is achieved by using a Convolutional Neural Network (UNET architecture).

Keywords: traffic light detection, image segmentation, machine learning, classification, convolutional neural networks

Procedia PDF Downloads 173

Authors: Mohammed Abdulridha Hamdan

Abstract:

To face the water crisis, understanding the role of human activities on nutrient concentrations in aquatic ecosystems needs more investigations to compare to extensively studies which have been carried out to understand these impacts on the water quality of different aquatic ecosystems. We hypothesized human activates on the catchments of Tigris river may change nutrient concentrations in water along the river. The results showed that phosphate concentration differed significantly among the studied sites due to distributed human activities, while nitrate concentration did not. Phosphate and nitrate concentrations were not affected by water temperature. We concluded that human activities on the surrounding landscapes could be more essential sources for nutrients of aquatic ecosystems than role of ongoing climate warming. Despite the role of warming in driving nutrients availability in aquatic ecosystems, our findings suggest to take the different activities on the surrounding catchments into account in the studies caring about the trophic status classification of aquatic ecosystems.

Keywords: nitrate, phosphate, anthropogenic, warming

Procedia PDF Downloads 81

Authors: Bensaid A., Mostephaoui T., Nedjai R.

Abstract:

A considerable area of Algerian lands are threatened by the phenomenon of wind erosion. For a long time, wind erosion and its associated harmful effects on the natural environment have posed a serious threat, especially in the arid regions of the country. In recent years, as a result of increases in the irrational exploitation of natural resources (fodder) and extensive land clearing, wind erosion has particularly accentuated. The extent of degradation in the arid region of the Algerian Mécheriadepartment generated a new situation characterized by the reduction of vegetation cover, the decrease of land productivity, as well as sand encroachment on urban development zones. In this study, we attempt to investigate the potential of remote sensing and geographic information systems for detecting the spatial dynamics of the ancient dune cords based on the numerical processing of PlanetScope PSB.SB sensors images by September 29, 2021. As a second step, we prospect the use of a multi-scale geographic object-based image analysis (GEOBIA) approach to segment the high spatial resolution images acquired on heterogeneous surfaces that vary according to human influence on the environment. We have used the fractal net evolution approach (FNEA) algorithm to segment images (Baatz&Schäpe, 2000). Multispectral data, a digital terrain model layer, ground truth data, a normalized difference vegetation index (NDVI) layer, and a first-order texture (entropy) layer were used to segment the multispectral images at three segmentation scales, with an emphasis on accurately delineating the boundaries and components of the sand accumulation areas (Dune, dunes fields, nebka, and barkhane). It is important to note that each auxiliary data contributed to improve the segmentation at different scales. The silted areas were classified using a nearest neighbor approach over the Naâma area using imagery. The classification of silted areas was successfully achieved over all study areas with an accuracy greater than 85%, although the results suggest that, overall, a higher degree of landscape heterogeneity may have a negative effect on segmentation and classification. Some areas suffered from the greatest over-segmentation and lowest mapping accuracy (Kappa: 0.79), which was partially attributed to confounding a greater proportion of mixed siltation classes from both sandy areas and bare ground patches. This research has demonstrated a technique based on very high-resolution images for mapping sanded and degraded areas using GEOBIA, which can be applied to the study of other lands in the steppe areas of the northern countries of the African continent.

Keywords: land development, GIS, sand dunes, segmentation, remote sensing

Procedia PDF Downloads 109

Authors: Pavan K. Rallabandi, Kailash C. Patidar

Abstract:

In this paper, we present an optimization technique or a learning algorithm using the hybrid architecture by combining the most popular sequence recognition models such as Recurrent Neural Networks (RNNs) and Hidden Markov models (HMMs). In order to improve the sequence or pattern recognition/ classification performance by applying a hybrid/neural symbolic approach, a gradient descent learning algorithm is developed using the Real Time Recurrent Learning of Recurrent Neural Network for processing the knowledge represented in trained Hidden Markov Models. The developed hybrid algorithm is implemented on automata theory as a sample test beds and the performance of the designed algorithm is demonstrated and evaluated on learning the deterministic finite state automata.

Keywords: hybrid systems, hidden markov models, recurrent neural networks, deterministic finite state automata

Procedia PDF Downloads 388

Authors: Nerella Srinivas

Abstract:

This study explores the significance of patent protection for artificial intelligence (AI) innovations in the pharmaceutical sector, emphasizing applications in drug discovery, personalized medicine, and clinical trial optimization. The challenges of patenting AI-driven inventions are outlined, focusing on the classification of algorithms as abstract ideas, meeting the non-obviousness standard, and issues around defining inventorship. The methodology includes examining case studies and existing patents, with an emphasis on how companies like Benevolent AI and Insilico Medicine have successfully secured patent rights. Findings demonstrate that a strategic approach to patent protection is essential, with particular attention to showcasing AI’s technical contributions to pharmaceutical advancements. Conclusively, the study underscores the critical role of understanding patent law and innovation strategies in leveraging intellectual property rights in the rapidly advancing field of AI-driven pharmaceuticals.

Keywords: artificial intelligence, pharmaceutical industry, patent protection, drug discovery, personalized medicine, clinical trials, intellectual property, non-obviousness

Procedia PDF Downloads 12

Authors: Feras Hanandeh, Majdi Shannag

Abstract:

This research studies the different factors that could affect the Faculty of Information Technology in Hashemite University students’ accumulative average. The research paper verifies the student information, background, their academic records, and how this information will affect the student to get high grades. The student information used in the study is extracted from the student’s academic records. The data mining tools and techniques are used to decide which attribute(s) will affect the student’s accumulative average. The results show that the most important factor which affects the students’ accumulative average is the student Acceptance Type. And we built a decision tree model and rules to determine how the student can get high grades in their courses. The overall accuracy of the model is 44% which is accepted rate.

Keywords: data mining, classification, extracting rules, decision tree

Procedia PDF Downloads 416

Authors: Maziyar Nouraee

Abstract:

Electronic markets in recent decades contribute remarkably in business transactions. Many organizations consider traditional ways of trade non-economical and therefore they do trade only through electronic markets. There are different categorizations of electronic markets functions. In one classification, functions of electronic markets are categorized into classes as information, transactions, and value added. In the present paper, effects of the three classes on the two major elements of the supply chain management are measured. The two elements are decrease in the product unit cost and reduction in response time to the customer. The results of the current research show that among nine minor elements related to the three classes of electronic markets functions, six factors and three factors influence on reduction of the product unit cost and reduction of response time to the customer, respectively.

Keywords: electronic commerce, electronic market, B2B trade, supply chain management

Procedia PDF Downloads 392

Authors: Chamseddine Barki, Salam Labidi, Hanen Boussi Rahmouni

Abstract:

In this article, we present the principle and suitable methodology for the design of a medical ontology that highlights the radiological and dosimetric knowledge, applied in diagnostic radiology and radiation-therapy. Our ontology, which we named «Onto.Rap», is the subject of radiation protection in medical and radiology centers by providing a standardized regulatory oversight. Thanks to its added values of knowledge-sharing, reuse and the ease of maintenance, this ontology tends to solve many problems. Of which we name the confusion between radiological procedures a practitioner might face while performing a patient radiological exam. Adding to it, the difficulties they might have in interpreting applicable patient radioprotection standards. Here, the ontology, thanks to its concepts simplification and expressiveness capabilities, can ensure an efficient classification of radiological procedures. It also provides an explicit representation of the relations between the different components of the studied concept. In fact, an ontology based-radioprotection expert system, when used in radiological center, could implement systematic radioprotection best practices during patient exam and a regulatory compliance service auditing afterwards.

Keywords: knowledge, ontology, radiation protection, radiology

Procedia PDF Downloads 312

Authors: N. W. U. D. Chathurani, Shlomo Geva, Vinod Chandran, Proboda Rajapaksha

Abstract:

Selecting an appropriate image representation is the most important factor in implementing an effective Content-Based Image Retrieval (CBIR) system. This paper presents a multi-feature fusion approach for efficient CBIR, based on the distance distribution of features and relative feature weights at the time of query processing. It is a simple yet effective approach, which is free from the effect of features' dimensions, ranges, internal feature normalization and the distance measure. This approach can easily be adopted in any feature combination to improve retrieval quality. The proposed approach is empirically evaluated using two benchmark datasets for image classification (a subset of the Corel dataset and Oliva and Torralba) and compared with existing approaches. The performance of the proposed approach is confirmed with the significantly improved performance in comparison with the independently evaluated baseline of the previously proposed feature fusion approaches.

Keywords: feature fusion, image retrieval, membership function, normalization

Procedia PDF Downloads 345

Authors: K. Badron, A. F. Ismail, A. L. Asnawi, N. F. A. Malik, S. Z. Abidin, S. Dzulkifly

Abstract:

The occurrences of precipitation, also commonly referred as rain, in the form of "convective" and "stratiform" have been identified to exist worldwide. In this study, the radar return echoes or known as reflectivity values acquired from radar scans have been exploited in the process of classifying the type of rain endured. The investigation use radar data from Malaysian Meteorology Department (MMD). It is possible to discriminate the types of rain experienced in tropical region by observing the vertical characteristics of the rain structure. .Heavy rain in tropical region profoundly affects radiowave signals, causing transmission interference and signal fading. Required wireless system fade margin depends on the type of rain. Information relating to the two mentioned types of rain is critical for the system engineers and researchers in their endeavour to improve the reliability of communication links. This paper highlights the quantification of percentage occurrences over one year period in 2009.

Keywords: stratiform, convective, tropical region, attenuation radar reflectivity

Procedia PDF Downloads 288