Computer Science Journal (AGH University of Science and Technology, Krakow)
Not a member yet
476 research outputs found
Sort by
A survey on multi-objective based parameter optimization for deep learning
No full textDeep learning models form one of the most powerful machine learning models for the extraction of important features. Most of the designs of deep neural models, i.e., the initialization of parameters, are still manually tuned. Hence, obtaining a model with high performance is exceedingly time-consuming and occasionally impossible. Optimizing the parameters of the deep networks, therefore, requires improved optimization algorithms with high convergence rates. The single objective-based optimization methods generally used are mostly time-consuming and do not guarantee optimum performance in allcases. Mathematical optimization problems containing multiple objective functions that must be optimized simultaneously fall under the category of multi-objective optimization sometimes referred to as Pareto optimization. Multi-objective optimization problems form one of the alternatives yet useful options for parameter optimization. However, this domain is a bit less explored. In this survey, we focus on exploring the effectiveness of multi-objective optimization strategies for parameter optimization in conjunction with deep neural networks. The case studies used in this study focus on how the twomethods are combined to provide valuable insights into the generation of predictions and analysis in multiple applications
Diacritic-aware yorùbá spell checker
Full text linkSpell checking and correction is still in infancy for Yorùbá language, and existing tools cannot be applied directly to address the problem because Yorùbá language requires extensive use of diacritics for marking phonemes and tone. We addressed this problem by collecting data from on-line sources and from optical character recognition of hard copy of books. The features relevant to spell checking and correction in this language that marks tones (and underdot) were identified through the review of existing spell checking solutions, analysis of the data collected and consultation with relevant Yorùbá Linguistics textbooks. A conceptual model was formulated as a parallel combination of a unigram language model and a language diacritic model to form a dictionary sub-model that is used by Error Detection and Candidate Generation modules. The candidate generation module was implemented as an inverse Levensthein edit-distance algorithm.
The system was evaluated using Detection Accuracy (calculated from Precision and Recall) and Suggestion Accuracy (SA) as metrics.Our experimental setups compared the performance of the component subsystems when used alone with the their combination into a unified model. The detection accuracies for each of the models were 93.23%, 94.10% and 95.01% respectively while their suggestion accuracies were 26.94%, 72.10% and 65.89% respectively. In relation to the size of training corpus, the unified model was able to achieve a increase of 1.83% in detection accuracy and 5.27% in suggestion accuracy for 70% increase in size of corpus. The results indicated that each of the sub-models in the dictionary played different roles while the increase in training data does not give a linear proportional increase in performance of the spell checker. The study also showed that spell checking a Yorùbá text was better when attention is paid to the diacritical aspect of the languag
Hybrid implementation of the fastICA algorithm for high-density EEG using the capabilities of the intel architecture and CUDA programming
Full text linkHigh-density electroencephalographic (EEG) systems are utilized in the study of the human brain and its underlying behaviors. However, working with EEG data requires a well-cleaned signal, which is often achieved through the use of independent component analysis (ICA) methods. The calculation time for these types of algorithms is the longer the more data we have. This article presents a hybrid implementation of the fastICA algorithm that uses parallel programming techniques (libraries and extensions of the Intel processors and CUDA programming), which results in a significant acceleration of execution time on selected architectures
Privacy preservation for transaction initiators: stronger key image ring signature and smart contract-based framework
Full text linkRecently, blockchain technology has garnered support. However, an attenuating factor to its global adoption in certain use cases is privacy-preservation owing to its inherent transparency. A widely explored cryptographic option to address this challenge has been ring signature which aside its privacy guarantee must be double spending resistant. In this paper, we identify and prove a catastrophic flaw for double-spending attack in a Lightweight Ring Signature scheme and proceed to construct a new, fortified commitment scheme using the signer’s entire private key. Subsequently, we compute a stronger key image to yield a double-spending-resistant signature scheme solidly backed by formal proof. Inherent in our solution is a novel, zero-knowledge-based, secured and cost-effective smart contract for public key aggregation. We test our solution on a private blockchain as well as Kovan testnet along with performance analysis attesting to efficiency and usability and make the code publicly available on GitHub
Explainable deep neural network based analysis on intrusion detection systems
Full text linkThe research on Intrusion Detection Systems (IDSs) have been increasing in recent years. Particularly, the research which are widely utilizing machine learning concepts, and it is proven that these concepts were effective with IDSs, particularly, deep neural network-based models enhanced the rate of detections of IDSs. At the same instance, the models are turning out to be very highly complex, users are unable to track down the explanations for the decisions made which indicates the necessity of identifying the explanations behind those decisions to ensure the interpretability of the framed model. In this aspect, the article deals with the proposed model that able to explain the obtained predictions. The proposed framework is a combination of a conventional intrusion detection system with the aid of a deep neural network and interpretability of the model predictions. The proposed model utilizes Shapley Additive Explanations (SHAP) that mixes with the local explainability as well as the global explainability for the enhancement of interpretations in the case of intrusion detection systems. The proposed model was implemented using the popular dataset, NSL-KDD, and the performance of the framework evaluated using accuracy, precision, recall, and F1-score. The accuracy of the framework is achieved by about 99.99%. The proposed framework able to identify the top 4 features using local explainability and the top 20 features using global explainability
Learning-free deep features for multispectral palm-print classification
Full text linkThe feature extraction step is a major and crucial step in analyzing and understanding raw data as it has a considerable impact on the system accuracy. Unfortunately, despite the very acceptable results obtained by many handcrafted methods, they can have difficulty representing the features in the case of large databases or with strongly correlated samples. In this context, we proposed a new, simple and lightweight method for deep feature extraction. Our method can be configured to produce four different deep features, each controlled to tune the system accuracy. We have evaluated the performance of our method using a multispectral palmprint based biometric system and the experimental results, using the CASIA database, have shown that our method has high accuracy compared to many current handcrafted feature extraction methods and many well known deep learning based methods
Transformation and classification of ordinal survey data
Full text linkCurrently, Machine Learning is being significantly used in almost all of the research domains. However, its applicability in survey research is still in its infancy. We in this paper, attempt to highlight the applicability of Machine Learning in survey research while working on two different aspects in parallel. First, we introduce a pattern-based transformation method for ordinal survey data. Our purpose behind developing such a transformation method is twofold. Our transformation facilitates easy interpretation of ordinal survey data and provides convenience while applying standard Machine Learning approaches. Second, we demonstrate the application of various classification techniques over real and transformed ordinal survey data and interpret their results in terms of their suitability in survey research. Our experimental results suggest that Machine Learning coupled with the Pattern Recognition paradigm has a tremendous scope in survey research
ARNLI: Arabic natural language inference entailment and contradiction detection
Full text linkNatural Language Inference (NLI) is a hot topic research in natural language processing, contradiction detection between sentences is a special case of NLI. This is considered a difficult NLP task which has a big influence when added as a component in many NLP applications, such as Question Answering Systems, text Summarization. Arabic Language is one of the most challenging low-resources languages in detecting contradictions due to its rich lexical, semantics ambiguity. We have created a dataset of more than 12k sentences and named ArNLI, that will be publicly available. Moreover, we have applied a new model inspired by Stanford contradiction detection proposed solutions on English language. We proposed an approach to detect contradictions between pairs of sentences in Arabic language using contradiction vector combined with language model vector as an input to machine learning model. We analyzed results of different traditional machine learning classifiers and compared their results on our created dataset (ArNLI) and on an automatic translation of both PHEME, SICK English datasets. Best results achieved using Random Forest classifier with an accuracy of 99%, 60%, 75% on PHEME, SICK and ArNLI respectively
Stacked denoising autoencoder based Parkinson’s disease classification using improved Pigeon-inspired optimization algorithm
No full textOne of the most common neurological conditions caused by gradual brain degeneration is Parkinson\u27s disease (PD). Although this neurological condition has no known treatment, early detection and therapy can help patients improve their quality of life. An essential patient\u27s health record is made of medical images used to control, manage, and treat diseases. However, in computer-based diagnostics, disease classification is a difficult task. To overcome this problem, this paper introduces a stacked denoising Autoencoder (SDA) for Parkinson\u27s disease classification. The main aim of this paper is to derive an optimal feature selection design for an effective PD classification. Improved Pigeon-Inspired Optimization (IPIO) algorithm is introduced to enhance the performance of the classifier. Thus, the classification result improved by the optimal features and also increased the sensitivity, accuracy, and specificity in the medical image diagnosis. The proposed scheme is implemented in PYTHON and compared with traditional feature selection models and other classification approaches. The experimental outcomes show that the proposed method yields a superior classification of PD than the current state-of-the-art metho
Hybrid variable neighborhood search for solving school bus-driver problem with resource constraints
Full text linkThe School Bus-Driver Problem with Resource Constraints (SBDP-RC) is an optimization problem with many practical applications. In the problem, the number of vehicles is prepared to pick a number of pupils, in which the total resource of all vehicles is less than a predefined value. The aim is to find a tour minimizing the sum of pupils’ waiting times. The problem is NP-hard in the general case. In many cases, reaching a feasible solution becomes an NP-hard problem. To solve the large-sized problem, a metaheuristic approach is a suitable approach. The first phase creates an initial solution by the construction heuristic based on Insertion Heuristic. After that, the post phase improves the solution by the General Variable Neighborhood Search (GVNS) with Random Neighborhood Search combined with Shaking Technique. The hybridization ensures the balance between exploitation and exploration. Therefore, the proposed algorithm can escape from local optimal solutions. The proposed metaheuristic algorithm is tested on a benchmark to show the efficiency of the algorithm. The results show that the algorithm receives good feasible solutions fast. Additionally, in many cases, better solutions can be found in comparison with the previous metaheuristic algorithms