SPEAKER IDENTIFICATION MODEL BASED ON DEEP NURAL NETWOKS

This study aims is to establish a small system of text-independent recognition of speakers for a relatively small group of speakers at a sound stage. The fascinating justification for the International Space Station (ISS) to detect if the astronauts are speaking at a specific time has influenced the difficulty. In this work, we employed Machine Learning Applications. Accordingly, we used the Direct Deep Neural Network (DNN)-based approach, in which the posterior opportunities of the output layer are utilized to determine the speaker’s presence. In line with the small footprint design objective, a simple DNN model with only sufficient hidden units or sufficient hidden units per layer was designed, thereby reducing the cost of parameters through intentional preparation to avoid the normal overfitting problem and optimize the algorithmic aspects, such as context-based training, activation functions, validation, and learning rate. Two commercially available databases, namely, TIMIT clean speech and HTIMIT multihandset communication database and TIMIT noise-added data framework, were tested for this reference model that we developed using four sound categories at three distinct signal-to-noise ratios. Briefly, we used a dynamic pruning method in which the conditions of all layers are simultaneously pruned, and the pruning mechanism is reassigned. The usefulness of this approach was evaluated on all the above contact databases

Pancreatic cancer survival prediction via inflammatory serum markers

Background For prognostic evaluation of pancreatic ductal adenocarcinoma (PDAC), the only well-established serum marker is carbohydrate antigen CA19-9. To improve the accuracy of survival prediction, we tested the efficacy of inflammatory serum markers. Methods A preoperative serum panel comprising 48 cytokines plus high-sensitivity CRP (hs-CRP) was analyzed in 173 stage I–III PDAC patients. Analysis of the effect of serum markers on survival utilized the Cox regression model, with the most promising cytokines chosen with the aid of the lasso method. We formed a reference model comprising age, gender, tumor stage, adjuvant chemotherapy status, and CA19-9 level. Our prognostic study model incorporated these data plus hs-CRP and the cytokines. We constructed time-dependent ROC curves and calculated an integrated time-averaged area under the curve (iAUC) for both models from 1 to 10 years after surgery. Results Hs-CRP and the cytokines CTACK, MIF, IL-1β, IL-3, GRO-α, M-CSF, and SCF, were our choices for the prognostic study model, in which the iAUC was 0.837 (95% CI 0.796–0.902), compared to the reference model’s 0.759 (95% CI 0.691–0.836, NS). These models divided the patients into two groups based on the maximum value of Youden’s index at 7.5 years. In our study model, 60th percentile survival times were 4.5 (95% CI 3.7–NA) years (predicted high-survival group, n = 34) and 1.3 (95% CI 1.0–1.7) years (predicted low-survival group, n = 128), log rank p < 0.001. By the reference model, the 60th percentile survival times were 2.8 (95% CI 2.1–4.4) years (predicted high-survival group, n = 44) and 1.3 (95% CI 1.0–1.7) years (predicted low-survival group, n = 118), log rank p < 0.001. Conclusion Hs-CRP and the seven cytokines added to the reference model including CA19-9 are potential prognostic factors for improved survival prediction for PDAC patients.

Models and Methods for Improving Performance of Wireless Computer Networks Based on the Decomposition of Lower Layers of the OSI Reference Model

The paper discusses models and methods for improving the performance of wireless computer networks built based on the decomposition of the lower levels of the OSI reference model. A method to improve the performance of networks is suggested, which functionally combines the physical and network layers, which improves its efficiency in marginal reception areas almost twice. A model of the block diagram of a device for improving data transmission quality in marginal reception areas or those with insufficient noise immunity is developed based on the so-called communication quality status monitoring, as well as a model of the block diagram of a wireless adaptive capacity reallocation router based on dynamic channels capacity reallocation, which allows adequately reallocating IS resources depending on traffic and user priority. Keywords— Bluetooth, FIFO discipline, IEEE 802.11, OSI/ISO reference model, wireless computer networks.

Specific Physical Ability Prediction in Youth Basketball Players According to Playing Position

This study investigated the hierarchical structure of physical characteristics in elite young (i.e., U17-U19) basketball players according to playing positions. In addition, their predictive value of physical characteristics was determined for the evaluation of players’ physical preparedness. Sixty elite male basketball players performed 13 standardized specific field tests in order to assess the explosive power of lower limbs, speed, and change-of-direction speed. They were divided into three groups according to playing positions (guard [n = 28], forward [n = 22], center [n = 10]). The basic characteristics of the tested sample were: age = 17.36 ± 1.04 years, body height = 192.80 ± 4.49 cm, body mass = 79.83 ± 6.94 kg, and basketball experience = 9.38 ± 2.10 years for guards; age = 18.00 ± 1.00 years, body height = 201.48 ± 3.14 cm, body mass = 90.93 ± 9.85 kg, and basketball experience = 9.93 ± 2.28 years for forwards; and age = 17.60 ± 1.43 years; body height = 207.20 ± 3.29 cm, body mass = 104.00 ± 9.64 kg, and basketball experience = 9.20 ± 1.62 years for centers. For all playing positions factor analysis extracted three factors, which cumulatively explained 76.87, 88.12 and 87.63% of variance, respectively. The assessed performance measures were defined as significant (p < 0.001), with regression models of physical performance index (PPINDEX). PPINDEX of guards = −6.860 + (0.932 × t-test) − (1.656 × Acceleration 15 m) − (0.020 × Countermovement jump); PPINDEX of forwards = −3.436 − (0.046 × Countermovement jump with arm swing) − (1.295 × Acceleration 15 m) + (0.582 × Control of dribbling); PPINDEX of centers = −4.126 + (0.604 × Control of dribbling) − (1.315 × Acceleration 15 m) − (0.037 × Sargent jump). A model for the evaluation of physical performance of young basketball players has been defined. In addition, this model could be used as a reference model for selection procedures, as well as to monitor the efficacy of applied training programmes within the short, medium and long-term periodization.

‘Thinking Ontologically’: Ontology Development as the Basis for the Systematic Analysis of a Corpus of Late Antique Aramaic Magic Texts

This article describes the process of developing an ontology of the domain of Jewish Babylonian Aramaic magic bowls and offers some reflections on its significance in the analysis of these materials. Examples are highlighted to illustrate where the work builds on existing conceptualisations of the domain in secondary literature and where magical and religious materials from the Ancient Near East might stimulate some specialised extension of the CIDOC Conceptual Reference Model (ICOM/CIDOC Documentation Standards Group, 2020). The analogy of ‘bridge building’ is offered as a way for humanities researchers to conceive of the work to produce ontologies of specific domains. This reflection is intended to capture the experience of ‘thinking ontologically’ about sources for the first time and of overcoming misconceptions about the nature and significance of this work.

Dual-Hybrid Modeling for Option Pricing of CSI 300ETF

The reasonable pricing of options can effectively help investors avoid risks and obtain benefits, which plays a very important role in the stability of the financial market. The traditional single option pricing model often fails to meet the ideal expectations due to its limited conditions. Combining an economic model with a deep learning model to establish a hybrid model provides a new method to improve the prediction accuracy of the pricing model. This includes the usage of real historical data of about 10,000 sets of CSI 300 ETF options from January to December 2020 for experimental analysis. Aiming at the prediction problem of CSI 300ETF option pricing, based on the importance of random forest features, the Convolutional Neural Network and Long Short-Term Memory model (CNN-LSTM) in deep learning is combined with a typical stochastic volatility Heston model and stochastic interests CIR model in parameter models. The dual hybrid pricing model of the call option and the put option of CSI 300ETF is established. The dual-hybrid model and the reference model are integrated with ridge regression to further improve the forecasting effect. The results show that the dual-hybrid pricing model proposed in this paper has high accuracy, and the prediction accuracy is tens to hundreds of times higher than the reference model; moreover, MSE can be as low as 0.0003. The article provides an alternative method for the pricing of financial derivatives.

Mechanical strength of highly preloaded bolts affected by the ovalization of an aircraft wheel

This work presented in this paper concerns the modeling of the tensile and bending behavior of bolts in an airplane wheel. The design of a very rigid airplane tire means that the airplane wheel must be separated into two parts. In order not to have a separation between the two parts, several bolts with high preload are used. The main objective of this work is to predict the mechanical behavior of this assembly in a preliminary design phase with geometrical and global mechanical data. To achieve this objective, a simplified semi-numerical 1D model is developed. The complex geometry of the wheels is modeled by axisymmetric elements, while beam elements define the geometries and mechanical behavior of the bolts. The model is improved in non-axisymmetric cases to include the ring effect due to the wheel ovalization. Different cases are simulated (inflation and rolling). For each load case, the most stressed fastener is examined. Then, a comparison between its static and fatigue stress results and those of the 3D finite element reference model considered is analyzed for the validation of the developed tool. The semi-numerical model is used in the preliminary design phase and permits the geometric and mechanical properties of the aircraft wheel and fasteners to be defined so as to find the best assembly configuration that prevents separation.

Ambient intelligence governance review: from service-oriented to self-service

The current generation of connected devices and the Internet of Things augment people’s capabilities through ambient intelligence. Ambient Intelligence (AmI) support systems contain applications consuming available services in the environment to serve users. A well-known design of these applications follows a service architecture style and implement artificial intelligence mechanisms to maintain an awareness of the context: The service architecture style enables the distribution of capabilities and facilitates interoperability. Intelligence and context-awareness provide an adaptation of the environment to improve the interaction. Smart objects in distributed deployments and the increasing machine awareness of devices and people context also lead us to architectures, including self-governed policies providing self-service. We have systematically reviewed and analyzed ambient system governance considering service-oriented architecture (SOA) as a reference model. We applied a systematic mapping process obtaining 198 papers for screening (out of 712 obtained after conducting searches in research databases). We then reviewed and categorized 68 papers related to 48 research projects selected by fulfilling ambient intelligence and SOA principles and concepts. This paper presents the result of our analysis, including the existing governance designs, the distribution of adopted characteristics, and the trend to incorporate service in the context-aware process. We also discuss the identified challenges and analyze research directions.

KnowMIS-ABSA: an overview and a reference model for applications of sentiment analysis and aspect-based sentiment analysis

The analysis of the opinions of customers and users has been always of great interest in supporting decision-making in many fields, especially in marketing. Sentiment analysis (SA) is the umbrella term for techniques and approaches that analyze user’s sentiments, emotions, opinions in text or other media. The need for a better understanding of these opinions paved the way to novel approaches that focus on the analysis of the sentiment related to specific features of a product, giving birth to the field of aspect-based sentiment analysis (ABSA). Although the increasing interest in this discipline, there is still confusion regarding the basic concepts of ABSA: terms like sentiment, affect, emotion, opinion, are used as synonyms while they represent different concepts. This often leads to an incorrect analysis of the users’ opinions.This work presents an overview of the state-of-the-art techniques and approaches for ABSA, highlighting the main critical issues related to current trends in this field. Following this analysis, a new reference model for SA and ABSA, namely the KnowMIS-ABSA model, is proposed. The model is grounded on the consideration that sentiment, affect, emotion and opinion are very different concepts and that it is profoundly wrong to use the same metric and the same technique to measure them. Accordingly, we argue that different tools and metrics should be adopted to measure each of the dimensions of an opinion. A qualitative case study, regarding product reviews, is proposed to motivate the advantages of the KnowMIS-ABSA model.

Design of an Artificial Intelligence of Things Based Indoor Planting Model for Mentha Spicata

In recent years, many large-scale plantings have become refined small-scale or home plantings. The rapid progress of agriculture technologies and information techniques enables people to control the growth of agricultural products well. Hence, this study proposes an Artificial Intelligence of Things (AIoT) based Plant Pot Design for planting edible mint in an office setting, which is called APPD. APPD is composed of intelligent gardens and a cloud-based service platform. An intelligent garden is deployed an Arduino with multiple sensors to monitor and control plant pots of the edible mint, Mentha spicata. The cloud-based service platform provides a Case-Based Reasoning (CBR) inference engine with a database for adjustment influence factors. This study discusses eight growing statuses of Mentha spicata with different illumination, photometric exposure, and moisture content, designed for an office environment. Evaluation results indicate that Mentha spicata with 16 h red–blue lighting and 50% moisture content makes a maximum 5% mint extract of the total weight of the mint leaves. Finally, APPD can be a reference model for researchers and engineers.