In 2016, Béziau introduces a restricted notion of paraconsistency, the so-called genuine paraconsistency. A logic is genuine paraconsistent if it rejects the laws $\varphi,\neg \varphi \vdash \psi$ and $\vdash \neg (\varphi \wedge \neg \varphi)$. In that paper, the author analyzes, among the three-valued logics, which of them satisfy this property. If we consider multiple-conclusion consequence relations, the dual properties of those above-mentioned are $\vdash \varphi, \neg \varphi$ and $\neg (\varphi \vee \neg \varphi) \vdash$. We call genuine paracomplete logics those rejecting (...) the mentioned properties. We present here an analysis of the three-valued genuine paracomplete logics. A very natural twist structures semantics for these logics is also found in a systematic way. This semantics produces automatically a simple and elegant Hilbert-style characterization for all these logics. Finally, we introduce the logic LGP which is genuine paracomplete is not genuine paraconsistent, not even paraconsistent and cannot be characterized by a single finite logical matrix. (shrink)

After proving, in a purely categorial way, that the inclusion functor |$\textrm {In}_{\textbf {Alg}(\varSigma )}$| from |$\textbf {Alg}(\varSigma )$|⁠, the category of many-sorted |$\varSigma $|-algebras, to |$\textbf {PAlg}(\varSigma )$|⁠, the category of many-sorted partial |$\varSigma $|-algebras, has a left adjoint |$\textbf {F}_{\varSigma }$|⁠, the (absolutely) free completion functor, we recall, in connection with the functor |$\textbf {F}_{\varSigma }$|⁠, the generalized recursion theorem of Schmidt, which we will also call the Schmidt construction. Next, we define a category |$\textbf {Cmpl}(\varSigma )$|⁠, of (...) |$\varSigma $|-completions, and prove that |$\textbf {F}_{\varSigma }$|⁠, labelled with its domain category and the unit of the adjunction of which it is a part, is a weakly initial object in it. Following this, we associate to an ordered pair |$(\boldsymbol {\alpha },f)$|⁠, where |$\boldsymbol {\alpha }=(K,\gamma,\alpha )$| is a morphism of |$\varSigma $|-completions from |${{\mathscr {F}}}=(\textbf {C},F,\eta )$| to |$\mathscr {G}= (\textbf {D},G,\rho )$| and |$f$| a homomorphism of |$\textbf {D}$| from the partial |$\varSigma $|-algebra |$\textbf {A}$| to the partial |$\varSigma $|-algebra |$\textbf {B}$|⁠, a homomorphism |$\varUpsilon ^{\mathscr {G},0}_{\boldsymbol {\alpha }}(f)\colon \textbf {Sch}_{\boldsymbol {\alpha }}(f)\longrightarrow \textbf {B}$|⁠. We then prove that there exists an endofunctor, |$\varUpsilon ^{\mathscr {G},0}_{\boldsymbol {\alpha }}$|⁠, of |$\textbf {Mor}_{\textrm {tw}}(\textbf {D})$|⁠, the twisted morphism category of |$\textbf {D}$|⁠, thus showing the naturalness of the previous construction. Afterwards, we prove that, for every |$\varSigma $|-completion |$\mathscr {G}=(\textbf {D},G,\rho )$|⁠, there exists a functor |$\varUpsilon ^{\mathscr {G}}$| from the comma category |$(\textbf {Cmpl}(\varSigma )\!\downarrow \!\mathscr {G})$| to |$\textbf {End}(\textbf {Mor}_{\textrm {tw}}(\textbf {D}))$|⁠, the category of endofunctors of |$\textbf {Mor}_{\textrm {tw}}(\textbf {D})$|⁠, such that |$\varUpsilon ^{\mathscr {G},0}$|⁠, the object mapping of |$\varUpsilon ^{\mathscr {G}}$|⁠, sends a morphism of |$\varSigma $|-completion of |$\textbf {Cmpl}(\varSigma )$| with codomain |$\mathscr {G}$|⁠, to the endofunctor |$\varUpsilon ^{\mathscr {G},0}_{\boldsymbol {\alpha }}$|⁠. (shrink)

We provide complete classifications of algebras of partial maps for a significant swathe of combinations of operations not previously classified. Our focus is the many subsidiary operations that arise in recent considerations of the ‘override’ and ‘update’ operations arising in specification languages. These other operations turn out to have an older pedigree: domain restriction, set subtraction and intersection. All signatures considered include domain restriction, at least as a term. Combinations of the operations are classified and given complete axiomatizations with and (...) without the presence of functional composition. Each classification is achieved by way of providing a concrete representation of the corresponding abstract algebras as partial maps acting on special kinds of filters determined with respect to various induced orders. In contrast to many negative results in the broader area, all of the considered combinations lead to finite axiomatizations. (shrink)

The Post, axled and Łukasiewicz–Moisil algebras are important lattices studied in algebraic logic. In this paper, we investigate a useful interpretation between these algebras and some rings. We give a term equivalence between Post algebras of order |$p$| and |$p$|-rings, |$p$| prime and lift this result to the axled Łukasiewicz–Moisil algebra |$L \cong B_s \times P$| and the ring |$\prod ^s F_2 \times \prod ^l F_p$|⁠, where |$B_s$| is a Boolean algebra of order |$2^s$|⁠, |$P$| a |$p$|-valued Post algebra of (...) order |$p^l$| and |$F_p$| is the prime field of order |$p$|⁠. (shrink)

The basis of the paper is a logic of analytical consequential implication, CI.0, which is known to be equivalent to the well-known modal system KT thanks to the definition A → B = df A ⥽ B ∧ Ξ (Α, Β), Ξ (Α, Β) being a symbol for what is called here Equimodality Property: (□A ≡ □B) ∧ (◊A ≡ ◊B). Extending CI.0 (=KT) with axioms and rules for the so-called circumstantial operator symbolized by *, one obtains a system CI.0*Eq (...) in whose language one can define an operator ↠ suitable to formalize context-dependent conditionals (so to counterfactual conditionals) via the definition (Def ↠) A ↠ B = df *A ⥽ B ∧ Ξ (Α, Β).. The central problem of the paper is to identify inside CI.0*Eq + Def ↠ a set of axioms yielding the fragment consisting of all and only theorems in which only truth-functional operators and the two operators → and ↠ occur. This system, here called CI.0, is introduced in §3. In view of the intended purpose, it is introduced a complete and tableau-decidable system KTw, which is an extension of KT with axioms for the so-called quasi-variables w(A), w(B)… Three translation functions among the three languages used in the paper are then introduced. The first, Tr, maps every wff of form *A into wffs of form w(A) ∧ A and translates theorems of CI.0*Eq into theorems of KTw. The second, t°, translates theorems of CI.0 into theorems of CI.0*Eq by applying Def↠. A third one, t, translates theorems of CI.0 into theorems of KTw. As a consequence, it follows that t°A = TrtA. In §4 it is proved that CI.0 is complete w.r.t. the class of CI.0-models of form where f is a selection function and R an access relation. It is then proved (i) that CI.0 models may be converted into KTw-models; (ii) that the truth-value of a proposition in a world of a CI.0-model is preserved in the same world of the derived KTw-model; (iii) that A is a CI.0-thesis iff its translation tA is KTw-thesis. It follows that, if A is not a CI.0-thesis, tA is not a KTw-thesis, so also that TrtA is not a CI.0*Eq-thesis. Since t°A = TrtA and t°A is a function built on Def↠, this proves that every t°-translation of a CI.0-wff that is a CI.0*Eq-theorem is a CI.0-theorem. Since the converse proposition is proved in §2, their conjunction establishes the desired result. In the final section, it is proved that CI.0 is tableau-decidable, that ↠ is not a trivial operator and that ↠ enjoys the positive and negative properties required in §1 for context-dependent conditionals. Some final remarks suggest that studying the relations between CI.0 and systems of classical conditional logic may be a promising line of investigation. (shrink)

Neural networks are widely used in systems of artificial intelligence, but due to their black box nature, they have so far evaded formal analysis to certify that they satisfy desirable properties, mainly when they perform critical tasks. In this work, we introduce methods for the formal analysis of reachability and robustness of neural networks that are modeled as rational McNaughton functions by, first, stating such properties in the language of Łukasiewicz infinitely-valued logic and, then, using the reasoning techniques of such (...) logical system. We also present a case study where we employ the proposed techniques in an actual neural network that we trained to predict whether it will rain tomorrow in Australia. (shrink)

We present a temporal logic of branching time with four primitive operators: |$\exists {\mathcal {C}}$| – it may change whether; |$\forall {\mathcal {C}} $| – it must change whether; |$\exists \Box $| – it may be endlessly unchangeable that; and |$\forall \Box $| – it must be endlessly unchangeable that. Semantically, operator |$\forall {\mathcal {C}}$| expresses a change in the logical value of the given formula in every state that may be an immediate successor of the one considered, while |$\exists (...) {\mathcal {C}}$| expresses a change in the logical value of the given formula in some state that is a possible immediate successor. |$\forall {\mathcal {C}} $| and |$\exists {\mathcal {C}}$| are not normal operators, they are not mutually definable and have unusual properties: |$\forall {\mathcal {C}} A\leftrightarrow \forall {\mathcal {C}} \neg A$| and |$\exists {\mathcal {C}} A \leftrightarrow \exists {\mathcal {C}} \neg A$| are axioms. Operators |$\exists \Box $| and |$\forall \Box $| express endless unchangeability in some and all paths, respectively. Our axiomatization contains two infinitary rules that allow one to obtain |$\exists \Box A$| from infinitely many combinations of formulas with |$A, \neg, \forall {\mathcal {C}}, \land $|⁠, and to get |$\forall \Box A$| from infinitely many combinations of formulas with |$A, \neg, \exists {\mathcal {C}}, \land $|⁠. We show that our formalism is complete by presenting a bridge between our logic and logic |$\textsf {UB}$|⁠, which is a fragment of |$\textsf {CTL}$| in which until operators does not occur. (shrink)

Affordable light detection and ranging sensors are becoming available for tasks such as simultaneous localization and mapping (SLAM) in robotics and autonomous driving; however, these sensors offer less quality data of lower resolution that hinders the performance of registration methods. The deep learning based approaches seem to be sensitive to these data flaws. Specifically, a state-of-the-art deep learning-based approach failed to produce meaningful results after several attempts to carry out transfer learning over a dataset collected indoors with one such affordable (...) sensors. The paper introduces a hybrid approach combining two well-established registration techniques, the iterative closest point algorithm and the normal distributions transform that achieves good performance on the SLAM task over the same dataset. (shrink)

Knowing the level of quality from which the context is no longer valuable in a Context-Aware Data Mining (CADM) system is an important information. The main goal of this research is to study the variations of the predictions in case of different levels of noise and missing context data in practical scenarios for predicting soil moisture. The research has been performed on two locations from the Transylvanian Plain, Romania and two locations from Canada. The values predicted for the soil moisture (...) were compared in mixed scenarios that vary the quantity of noise and missing context data. The studied behavior was performed using Deep Learning, Decision Tree and Gradient Boosted Tree machine learning algorithms. It has been shown that when using the air temperature as context for predicting soil moisture, variations of noise and missing data do not influence the results proportionally with the levels of noise and missing data applied. Also, Gradient Boosted Tree algorithm proves to be the best algorithm from the ones studied, to be considered when predicting soil moisture with the CADM approach. (shrink)

The results of evolutionary algorithms depend on population diversity that normally decreases by increasing the selection pressure from generation to generation. Usually, this can lead the evolution process to get stuck in local optima. This study is focused on mechanisms to avoid this undesired phenomenon by introducing parallel self-adapted differential evolution that decomposes a monolithic population into more variable-sized sub-populations and combining this with the characteristics of evolutionary multi-agent systems into a hybrid algorithm. The proposed hybrid algorithm operates with individuals (...) having some characteristics of agents, e.g. they act autonomously by selecting actions, with which they affect the state of the environment. Additionally, this algorithm incorporates two additional mechanisms: ageing and adaptive population growth, which help the individuals by decision-making. The proposed parallel differential evolution was applied to the CEC’18 benchmark function suite, while the produced results were compared with some traditional stochastic nature-inspired population-based and state-of-the-art algorithms. (shrink)

The price of copper is unstable but it is considered an important indicator of the global economy. Changes in the price of copper point to higher global growth or an impending recession. In this work, the forecasting of the spot prices of copper from the New York Commodity Exchange is studied using a machine learning method, support vector regression coupled with different model schemas (recursive, direct and hybrid multi-step). Using these techniques, three different time series analyses are built and its (...) performance are compared. The numerical results show that the hybrid direct-recursive obtains the best results. (shrink)

The travelling salesman problem (TSP) belongs to the class of NP-hard problems, in which an optimal solution to the problem cannot be obtained within a reasonable computational time for large-sized problems. To address TSP, we propose a hybrid algorithm, called GA-TCTIA-LBSA, in which a genetic algorithm (GA), tour construction and tour improvement algorithms (TCTIAs) and a list-based simulated annealing (LBSA) algorithm are used. The TCTIAs are introduced to generate a first population, and after that, a search is continued with the (...) GA. The problem of premature convergence of the GA to local optimum is tackled by a method called social disaster technique. Afterwards, the LBSA is applied to generate a new population based on one of two proposed operators called packing and judgement day. The proposed algorithm is implemented in the MATLAB environment, and its two variants, called GA-TCTIA-LBSA packing and GA-TCTIA-LBSA judgement day, are tested on symmetric and asymmetric instances from TSPLIB. The overall results demonstrate that the proposed GA-TCTIA-LBSAs offer promising results, particularly for small-sized instances. (shrink)

Processing data streams arriving at high speed requires the development of models that can provide fast and accurate predictions. Although deep neural networks are the state-of-the-art for many machine learning tasks, their performance in real-time data streaming scenarios is a research area that has not yet been fully addressed. Nevertheless, much effort has been put into the adaption of complex deep learning (DL) models to streaming tasks by reducing the processing time. The design of the asynchronous dual-pipeline DL framework allows (...) making predictions of incoming instances and updating the model simultaneously, using two separate layers. The aim of this work is to assess the performance of different types of DL architectures for data streaming classification using this framework. We evaluate models such as multi-layer perceptrons, recurrent, convolutional and temporal convolutional neural networks over several time series datasets that are simulated as streams at different speeds. In addition, we evaluate how the different architectures react to concept drifts typically found in evolving data streams. The obtained results indicate that convolutional architectures achieve a higher performance in terms of accuracy and efficiency, but are also the most sensitive to concept drifts. (shrink)

Growth modelling is of interest to scientists in various disciplines. In our article, we will collect 17 models designed for growth modelling, appraise these models and contribute to the discussion of their applicability. The merit of the paper lies in studying the convergence properties of nonlinear regression in selected models. Our studies will be performed mainly concerning the quality of the obtained estimates, which are closely related to the intrinsic curvature of the model according to Bates and Watts. This curvature (...) determines the size of the linearization domains. Only if the initial solution is in this domain, then the convergence of the estimate in nonlinear regression is guaranteed. The primary goal is to design a methodology for selecting a growth model. We will demonstrate fruitfulness of our methodology on the weight measurements of 10 calves under 25 months of age from cowsheds in the village Záluží in the Czech Republic. Estimated parameters of growth curves, sizes of linearization domain, calculated residues and coefficients of determination indices will be the subject of discussion. (shrink)

The Air Quality Index (AQI) shows the state of air pollution in a unique and more understandable way. This work aims to forecast the AQI in Algeciras (Spain) 8 hours in advance. The AQI is calculated indirectly through the predicted concentrations of five pollutants (O3, NO2, CO, SO2 and PM10) to achieve this goal. Artificial neural networks (ANNs), sequence-to-sequence long short-term memory networks (LSTMs) and a newly proposed method combing a rolling window with the latter (LSTMNA) are employed as the (...) forecasting techniques. Besides, two input approaches are evaluated: using only the data from the own time series of the pollutant in the first case or adding exogenous variables in the second one. Several window sizes are employed (24, 28 and 72 hours) with ANNs and LSTMNAs. Additionally, several feature ranking methods are applied in the exogenous approach to select the most relevant lagged variables to feed the models. Results show how the proposed exogenous approach increases the performance of the prediction models. Besides, the newly proposed method LSTMNA provides the best performances in most of the cases evaluated. Hence, it constitutes an exciting alternative to standard LSTMs and ANNs to predict pollutants concentrations and, consequently, the AQI. (shrink)

The harmful consequences of fossil fuels use has resulted in the promotion of clean and renewable energies. During the past decades, green technologies have experienced a strong development, paying especial attention to wind energy, that covers a significant share of the electric energy demand. In this context, the main efforts are focused on the optimization of wind generator facilities, not only in the mechanic design but also in the energy management. Then, the present work deals with the prediction of the (...) energy generated in a small wind turbine placed in a bioclimatic house located on the north west region of Spain. This includes an analysis of the characteristics of the atmospheric variables registered during the turbine operation for a period of one year and an exploratory examination of a range of regression techniques in order to assess the suitability of using the registered information to predict the installation’s power generation levels on the short term. The study detailed in this work proves that this objective is an attainable one with a good degree of accuracy. (shrink)

We present a pocket-size densely connected convolutional network (DenseNet) directed to classification of size-normalized colour images according to varieties of grapes captured in those images. We compare the DenseNet with three established small-size networks in terms of performance, inference time and model size. We propose a data augmentation that we use in training the networks. We train and evaluate the networks on in-field images. The trained networks distinguish between seven grapevine varieties and background, where four and three varieties, respectively, are (...) of red and green grapes. Compared to the established networks, the DenseNet is characterized by near state-of-the-art performance, short inference time and minimal model size. All these aspects qualify the network for real-time, mobile and edge computing applications. The DenseNet opens possibilities for constructing affordable selective harvesters in accordance with agriculture 4.0. (shrink)

In the present work, an artificial neural network (ANN) will be developed to estimate the relative rotation and translation of the autonomous mobile robot (AMR). The ANN will work as an iterative closed point, which is commonly used with the singular value decomposition algorithm. This development will provide better resolution for a relative positioning technique that is essential for the AMR localization. The ANN requires a specific architecture, although in the current work a neural architecture search will be adapted to (...) select the best ANN for estimating the relative motion. At the end, these ANNs will be compared with conventional algorithms to check the good performance of adopting an intelligent method for relative positioning estimation. (shrink)

The assembly of products or components by operators in industries is a complex task with recurring problems. In these processes, operators often make errors that can lead to defective products. Therefore, they need to be inspected later to verify their correct assembly. The main problems are caused by several reasons including high employee turnover, lack of experience in manufacturing specific products or confusion in interpreting instructions to assemble similar components. In this paper, a novel structured language aimed to describe the (...) required actions to manufacture a product in industrial assembly environments is presented. The main contribution is to provide a formal language to feed automatic systems that can verify through visual control whether the actions performed by the operator are carried out following the standard described by this language. In general, the use of this formal language will allow to minimize the negative impact of errors during assembly and to reduce waste in many forms in the industry. (shrink)

In this study, a hybrid model based on intelligent techniques is developed to predict the active power generated in a bioclimatic house by a low power wind turbine. Contrary to other researches that predict the generated power taking into account the speed and the direction of the wind, the model developed in this paper only uses the speed of the wind, measured mainly in a weather station from the government meteorological agency (MeteoGalicia). The wind speed is measured at different heights, (...) against the usual measurements in others researches, which uses the wind speed and the direction measured in a weather station on the wind turbine nacelle. The prediction is performed 30 minutes ahead, what ensures that the Building Management System knows the energy generated by the low power wind turbine 30 minutes before, and it can adapt the consumption of different equipment in the house to optimize the power use. The main objective is to allow the Building Management System to optimize the uses of energy, taking into account the predicted amount of energy that will be produced and the energy consumed in the house. The developed model uses a hybrid topology with four clusters to improve the prediction, achieving an error lower than 6.5% for Mean Absolute Error measured in a final test. To perform this test, part of the original dataset was isolated from the beginning of the training process to check the model with a dataset that is not used before, simulating the model as it is receiving new data. (shrink)

In the field of cybernetic systems and more specifically in robotics, one of the fundamental objectives is the detection of anomalies in order to minimize loss of time. Following this idea, this paper proposes the implementation of a Hybrid Intelligent System in four steps to impute the missing values, by combining clustering and regression techniques, followed by balancing and classification tasks. This system applies regression models to each one of the clusters built on the instances of data set. Subsequently, a (...) variety of balancing techniques are applied to improve the classifier’s ability to discern whether it is in an error or a normal state. These techniques support to obtain better classification ratios in which a robot is close to error and allow us to bring the behavior back to a normal state. The experimentation is performed using a modern and public data set, which has been extracted from a component-based robotic system, in which different anomalies are induced by software in their components. (shrink)

Matrices that cannot be handled using conventional clustering, regression or classification methods are often found in every big data research area. In particular, datasets with thousands or millions of rows and less than a hundred columns regularly appear in biological so-called omic problems. The effectiveness of conventional data analysis approaches is hampered by this matrix structure, which necessitates some means of reduction. An evolutionary method called PreCLAS is presented in this article. Its main objective is to find a submatrix with (...) fewer rows that exhibits some group structure. Three stages of experiments were performed. First, a benchmark dataset was used to assess the correct functionality of the method for clustering purposes. Then, a microarray gene expression data matrix was used to analyze the method’s performance in a simple classification scenario, where differential expression was carried out. Finally, several classification methods were compared in terms of classification accuracy using an RNA-seq gene expression dataset. Experiments showed that the new evolutionary technique significantly reduces the number of rows in the matrix and intelligently performs unsupervised row selection, improving classification and clustering methods. (shrink)

In this paper incomplete data sets, or data sets with missing attribute values, have three interpretations, lost values, attribute-concept values and ‘do not care’ conditions. Additionally, the process of data mining is based on two types of probabilistic approximations, global and saturated. We present results of experiments on mining incomplete data sets using six approaches, combining three interpretations of missing attribute values with two types of probabilistic approximations. We compare our six approaches, using the error rate computed as a result (...) of ten-fold cross validation as a criterion of quality. We show that for some data sets the error rate is significantly smaller (5% level of significance) for lost values, for some data sets the smaller error rate is associated with attribute-concept values, and sometimes with ‘do not care’ conditions. Again, for some approaches the error rate is significantly smaller for saturated probabilistic approximations than for global probabilistic approximations, while for some approaches it is the other way around. Thus, for an incomplete data set, the best approach to data mining should be chosen by trying all six approaches. (shrink)

There are many real-world applications like healthcare systems, job monitoring, well-being and personal fitness tracking, monitoring of elderly and frail people, assessment of rehabilitation and follow-up treatments, affording Fall Detection (FD) and ADL (Activity of Daily Living) identification, separately or even at a time. However, the two main drawbacks of these solutions are that most of the times, the devices deployed are obtrusive (devices worn on not quite common parts of the body like neck, waist and ankle) and the poor (...) battery life. Thus, this work proposes a low-power classification algorithm based on an Ensemble of KNN and K-Means algorithms (EKMeans) to identify Falls and High-Intensity ADL events such as running, jogging and climbing up stairs. The input of EKMeans are triaxial accelerometer data gathered from wrist-wearable devices. The proposal will be validated on the Fall&ADL publicly available datasets UMAFall, UCIFall and FallAllD, considering two kinds of activity labelling: Two-Class and Multi-Class. An exhaustive comparative study between our proposal, and the baseline algorithms KNN and a feed-forward Neural Network (NN) is deployed, where EKMeans outperformed clearly the Specificity (ADL classification) of the KNN and NN for the three datasets. Finally, a comparative battery consumption study has been included deploying the analyzed algorithms in a WearOS smartwatch, where EKMeans drains the battery from 100% to 0% in 27.45 hours, saving 5% and 21% concerning KNN and NN, respectively. Keywords: Human Activity Recognition, ADL Identification, Fall Detection TS Clustering, TS Classification, Wearable Devices, Low-Power HAR. (shrink)

This paper addresses a variant of the job shop scheduling problem with total tardiness minimization where task durations and due dates are uncertain. This uncertainty is modelled with intervals. Different ranking methods for intervals are considered and embedded into a genetic algorithm. A new robustness measure is proposed to compare the different ranking methods and assess their capacity to predict ‘expected delays’ of jobs. Experimental results show that dealing with uncertainty during the optimization process yields more robust solutions. A sensitivity (...) analysis also shows that the robustness of the solutions given by the solving method increases when the uncertainty grows. (shrink)

Inception module is one of the most used variants in convolutional neural networks. It has a large portfolio of success cases in computer vision. In the past years, diverse inception flavours, differing in the number of branches, the size and the number of the kernels, have appeared in the scientific literature. They are proposed based on the expertise of the practitioners without any optimization process. In this work, an implementation of population-based incremental learning is proposed for automatic optimization of the (...) hyperparameters of the inception module. This hyperparameters optimization undertakes classification of the MNIST database of handwritten digit images. This problem is widely used as a benchmark in classification, and therefore, the learned best configurations for the Inception module will be of wide use in the deep learning community. In order to reduce the carbon footprint of the optimization process, policies for reducing the redundant evaluations have been undertaken. As a consequence of this work, an evaluation of configurations of the inception module and a mechanism for optimizing hyperparameters in deep learning architectures are stated. (shrink)

This work presents a novel approach to forecast streaming big time series based on nearest similar patterns. This approach combines a clustering algorithm with a classifier and the nearest neighbours algorithm. It presents two separate stages: offline and online. The offline phase is for training and finding the best models for clustering, classification and the nearest neighbours algorithm. The online phase is to predict big time series in real time. In the offline phase, data are divided into clusters and a (...) forecasting model based on the nearest neighbours is trained for each cluster. In addition, a classifier is trained using the cluster assignments previously generated by the clustering algorithm. In the online phase, the classifier predicts the cluster label of an instance, and the proper nearest neighbours model according to the predicted cluster label is applied to obtain the final prediction using the similar patterns. The algorithm is able to be updated incrementally for online learning from data streams. Results are reported using electricity consumption with a granularity of |$10$| minutes for 4-hour-ahead forecasting and compared with well-known online benchmark learners, showing a remarkable improvement in prediction accuracy. (shrink)

Topic modelling is a popular unsupervised method for text processing that provides interpretable document representation. One of the most high-level approaches is additively regularized topic models (ARTM). This method features better quality than other methods due to its flexibility and advanced regularization abilities. However, it is challenging to find an optimal learning strategy to create high-quality topics because a user needs to select the regularizers with their values and determine the order of application. Moreover, it may require many real runs (...) or model training which makes this task time consuming. At the current moment, there is a lack of research on parameter optimization for ARTM-based models. Our work proposes an approach that formalizes the learning strategy into a vector of parameters which can be solved with evolutionary approach. We also propose a surrogate-based modification which utilizes machine learning methods that makes the approach for parameters search time efficient. We investigate different optimization algorithms (evolutionary and Bayesian) and their modifications with surrogates in application to topic modelling optimization using the proposed learning strategy approach. An experimental study conducted on English and Russian datasets indicates that the proposed approaches are able to find high-quality parameter solutions for ARTM and substantially reduce the execution time of the search. (shrink)

Entity alignment is the task of integrating heterogeneous knowledge among different knowledge graphs (KGs). KG is a popular way of storing facts about real-world entities. Unfortunately, a very limited number of the entities stored in different KGs are aligned. This paper presents an embedding-based entity alignment method that finds entity alignment by measuring the similarities between entity embeddings. Existing methods mainly focus on the relational structures and attributes information for the alignment process. Such methods fail while the entities have a (...) fewer number of attributes or when the relational structure could not capture the meaningful representation of the entities. To address this problem, we propose EASAE, an entity alignment method using summary and attribute embeddings. We exploit the entity summary information available in KGs for entities’ summary embedding. To learn the semantics of the entity summary, we employ bidirectional encoder representations from transformers (BERT). Our model learns the representations of entities by using relational triples, attribute triples and summary as well. We perform experiments on real-world datasets, and the results indicate that the proposed approach outperformed the state-of-the-art models for entity alignment. (shrink)

Headache disorders can be considered as the predominant neurological condition. In the field of neurological diseases, migraine was estimated to cost a total of €27 billion per year for the loss through reduced work productivity in the European Community. Medical data and information in turn provide knowledge based on which physicians make scientific decisions for diagnosis and treatments. It is, therefore, very useful to create diagnostic tools to help physicians make better decisions. This paper is focused on a new hybrid (...) clustering system combining analytic hierarchy process and weighted fuzzy c-means clustering method for diagnosing children with primary headache disorder. The proposed three-stage hybrid diagnosing system is tested on data set collected from hospitalized children in the Clinical Centre of Vojvodina, Novi Sad, Serbia. (shrink)

The usage of new and progressive technologies brings with it new types of security threats and security incidents. Their number is constantly growing.The current trend is to move from reactive to proactive activities. For this reason, the organization should be aware of the current security situation, including the forecasting of the future state. The main goal of organizations, especially their security operation centres, is to handle events, identify potential security incidents, and effectively forecast the network security situation awareness (NSSA). In (...) this paper, we focus on increasing the efficiency of utilization of this part of cybersecurity. The paper’s main aim is to compare selected statistical models and models based on neural networks to find out which models are more suitable for NSSA forecasting. Based on the analysis provided in this paper, neural network methods prove a more accurate alternative than classical statistical prediction models in NSSA forecasting. In addition, the paper analyses the selection criteria and suitability of time series, which do not only reflect information about the total number of security events but represent a category of security event (e.g. recon scanning), port or protocol. (shrink)

In this research work a novel two-step system for anomaly detection is presented and tested over several real datasets. In the first step the novel Exploratory Projection Pursuit, Beta Hebbian Learning algorithm, is applied over each dataset, either to reduce the dimensionality of the original dataset or to face nonlinear datasets by generating a new subspace of the original dataset with lower, or even higher, dimensionality selecting the right activation function. Finally, in the second step Principal Component Analysis anomaly detection (...) is applied to the new subspace to detect the anomalies and improve its classification capabilities. This new approach has been tested over several different real datasets, in terms of number of variables, number of samples and number of anomalies. In almost all cases, the novel approach obtained better results in terms of area under the curve with similar standard deviation values. In case of computational cost, this improvement is only remarkable when complexity of the dataset in terms of number of variables is high. (shrink)

It is already known that unifiable formulas in normal modal logic |$\textbf {K}+\square ^{2}\bot $| are either finitary or unitary and unifiable formulas in normal modal logic |$\textbf {Alt}_{1}+\square ^{2}\bot $| are unitary. In this paper, we prove that for all |$d{\geq }3$|⁠, unifiable formulas in normal modal logic |$\textbf {K}+\square ^{d}\bot $| are either finitary or unitary and unifiable formulas in normal modal logic |$\textbf {Alt}_{1}+\square ^{d}\bot $| are unitary.

In a multiagent and multi-cultural world, the fine-grained analysis of agents’ dynamic behaviour, i.e. of their activities, is essential. Dynamic activities are actions that are characterized by an agent who executes the action and by other participants of the action. Wh-questions on the participants of the actions pose a difficult particular challenge because the variability of the types of possible answers to such questions is huge. To deal with the problem, we propose the analysis and classification of Wh-questions apt for (...) agents’ communication in a multiagent system (MAS). Our proposal of such a system consists of agents who communicate with their fellow agents by messaging so that each autonomous agent, though resource-bounded, can make less or more rational decisions to meet its own and collective goals. In addition, by communicating with other fellow agents and their environment, agents can learn new concepts and enrich their ontology so that their behaviour is dynamic. We aim to make a general proposal of the system so that the ‘envelope’ of agents’ messages can be formalized in any MAS standard, be it The Foundation for Intelligent Physical Agents - Agent Communication Language (FIPA-ACL) or Knowledge Query and Manipulation Language (KQML). Yet, the content of messages is encoded in a formalized natural language. To this end, we apply Transparent Intensional Logic (TIL) with its procedural semantics which is particularly apt for a fine-grained analysis in which all the semantically salient features of natural language can be plausibly formalized. In this paper, we concentrate on analysing the content of query messages, particularly the content of those that encode Wh-questions and the answers to them. We also summarize TIL deduction system that makes it possible to answer such questions in an intelligent way. Linguists distinguish several subtypes of Wh-questions. Though the linguistic classification is helpful, it is not always suitable for agents’ communication and reasoning. We need the classification based on a logical analysis of Wh-questions so that the agents can infer possible answers to such questions rather than only looking for them by keywords. This paper aims to apply an appropriate classification of the logical types of Wh-questions and the analysis of such questions; we concentrate in particular on questions concerning the participants of activities. The application of these results to the analysis of processes and events based on verb valency frames is another novelty of the paper. (shrink)

Temporal justification logic is a new family of temporal logics of knowledge in which the knowledge of agents is modelled using a justification logic. In this paper, we present various temporal justification logics involving both past and future time modalities. We combine Artemov’s logic of proofs with linear temporal logic with past, and we also investigate several principles describing the interaction of justification and time. We present two kinds of semantics for our temporal justification logics, one based on interpreted systems (...) and Fitting models and the other based on Mkrtychev models, and further, we establish soundness and completeness. We show that the internalization property holds in some of the temporal justification logics. We further investigate the two well-known epistemic-temporal notions of no forgetting and no learning in the framework of justification logics. Finally, we present temporal justification logics that avoid the logical omniscience problem. (shrink)

Weak filters were introduced by K. Schlechta in the ’90s with the aim of interpreting defaults via a generalized ‘most’ quantifier in first-order logic. They arguably represent the largest class of structures that qualify as a ‘collection of large subsets’ of a given index set |$I$|⁠, in the sense that it is difficult to think of a weaker, but still plausible, definition of the concept. The notion of weak ultrafilter naturally emerges and has been used in epistemic logic and other (...) knowledge representation (KR) applications. We provide a comprehensive exposition of weak filters and ultrafilters, comparing them with their classical counterparts that have found very important applications in logic, set theory and topology. Weak (ultra)filters capture the ‘majorities’ of social choice theory (which coincide with the commonsense understanding of a ‘large subset’) and in that respect, they outperform classical (ultra)filters in their role as ‘collections of large subsets’. Yet, they lack some of the elegant properties of classical (ultra)filters that make them an almost perfect match for logical theories. We investigate the extent to which some important classical results carry through in this new setting, and we focus on genuinely weak filters and genuinely weak ultrafilters. For weak ultrafilters, we proceed to provide concrete examples, answer questions of existence and characterize their construction. The class of weak (ultra)filters represents a genuine contribution of KR to set theory that might be of some interest to set theorists too and we initiate this study by providing a glimpse on natural set-theoretic questions. (shrink)

The interpretability logic ILP is the interpretability logic of all sufficiently strong |$\varSigma _1$|-sound finitely axiomatised theories, such as the Gödel-Bernays set theory. The interpretability logic IL is a strict subset of the intersection of the interpretability logics of all so-called reasonable theories, IL(All). It is known that both ILP and ILW are decidable, however their complexity has not been resolved previously. In [10] it was shown that the basic interpretability logic IL is PSPACE-complete. Here we prove the same for (...) ILP and ILW. (shrink)

Classical logic of formal provability includes Löb’s theorem, but not reflection. In contrast, intuitions about the inferential behavior of informal provability (in informal mathematics) seem to invalidate Löb’s theorem and validate reflection (after all, the intuition is, whatever mathematicians prove holds!). We employ a non-deterministic many-valued semantics and develop a modal logic T-BAT of an informal provability operator, which indeed does validate reflection and invalidates Löb’s theorem. We study its properties and its relation to known provability-related paradoxical arguments. We also (...) argue that T-BAT is a fairly sensible candidate for a formal logic of informal provability. (shrink)

In the originally published version of this manuscript, several errors were noted and listed in this corrigendum. The author apologises for these inaccuracies.

In this article, I consider the positive logic of Boolean groups (i.e. Abelian groups where every non-identity element has order 2), where these are taken as frames for an operational semantics à la Urquhart. I call this logic BG. It is shown that the logic over the smallest nontrivial Boolean group, taken as a frame, is identical to the positive fragment of a quasi-relevance logic that was developed by Robles and Méndez (an extension of this result where negation is included (...) is also discussed). It is proved that BG satisfies the variable sharing property and is Halldén complete, while failing to satisfy the disjunction property. Sound and complete subscripted tableaux are presented for BG. An axiomatization (in the positive language extended with fusion) is presented, which is sound with respect to BG and, with respect to a related ternary relational semantics, complete; the problem of identifying a complete axiomatization for BG itself is left open. Connections between BG and the quasi-relevance logic KR are also discussed. (shrink)

In the present paper, we develop the algorithmic correspondence theory for hybrid logic with binder |$\mathcal {H}(@, \downarrow )$|⁠. We define the class of Sahlqvist inequalities for |$\mathcal {H}(@, \downarrow )$|⁠, and each inequality of which is shown to have a first-order frame correspondent effectively computable by an algorithm |$\textsf {ALBA}^{\downarrow }$|⁠.