A New Method for Sentiment Analysis Using Contextual Auto-Encoders

Sentiment analysis, a hot research topic, presents new challenges for understanding users’ opinions and judgments expressed online. They aim to classify the subjective texts by assigning them a polarity label. In this paper, we introduce a novel machine learning framework using auto-encoders network to predict the sentiment polarity label at the word level and the sentence level. Inspired by the dimensionality reduction and the feature extraction capabilities of the auto-encoders, we propose a new model for distributed word vector representation “PMI-SA” using as input pointwise-mutual-information “PMI” word vectors. The resulted continuous word vectors are combined to represent a sentence. An unsupervised sentence embedding method, called Contextual Recursive Auto-Encoders “CoRAE”, is also developed for learning sentence representation. Indeed, CoRAE follows the basic idea of the recursive auto-encoders to deeply compose the vectors of words constituting the sentence, but without relying on any syntactic parse tree. The CoRAE model consists in combining recursively each word with its context words (neighbors’ words: previous and next) by considering the word order. A support vector machine classifier with fine-tuning technique is also used to show that our deep compositional representation model CoRAE improves significantly the accuracy of sentiment analysis task. Experimental results demonstrate that CoRAE remarkably outperforms several competitive baseline methods on two databases, namely, Sanders twitter corpus and Facebook comments corpus. The CoRAE model achieves an efficiency of 83.28% with the Facebook dataset and 97.57% with the Sanders dataset.     

Adaptive discrete-time sliding-mode control of nonlinear systems described by Wiener models

In this paper, we propose an adaptive control scheme that can be applied to nonlinear systems with unknown parameters. The considered class of nonlinear systems is described by the block-oriented models, specifically, the Wiener models. These models consist of dynamic linear blocks in series with static nonlinear blocks. The proposed adaptive control method is based on the inverse of the nonlinear function block and on the discrete-time sliding-mode controller. The parameters adaptation are performed using a new recursive parametric estimation algorithm. This algorithm is developed using the adjustable model method and the least squares technique. A recursive least squares (RLS) algorithm is used to estimate the inverse nonlinear function. A time-varying gain is proposed, in the discrete-time sliding mode controller, to reduce the chattering problem. The stability of the closed-loop nonlinear system, with the proposed adaptive control scheme, has been proved. An application to a pH neutralisation process has been carried out and the simulation results clearly show the effectiveness of the proposed adaptive control scheme.     

A recursive hierarchical parametric estimation algorithm for nonlinear systems described by Wiener‐Hammerstein models

In this paper, a recursive hierarchical parametric estimation (RHPE) algorithm is proposed for stochastic nonlinear systems which can be described by Wiener‐Hammerstein (W‐H) mathematical models. The formulation of parameters estimation problem is based on the prediction error approach and the gradient techniques. The convergence analysis of the developed RHPE algorithm is derived using stochastic gradient‐based theory. Wiener‐Hammerstein hydraulic process is treated to prove the efficiency of the proposed approach.     

Arabic stop consonants characterisation and classification using the normalized energy in frequency bands

In general, speech is made with sequences of consonants (fricatives, nasals and stops), vowels and glides. The classification of the stop consonants remains one of the most challenging problems in speech recognition. In this paper, we propose a new approach based on the normalized energy in frequency bands in the release and closure phases in order to characterize and classify the Arabic stop consonants (/b/, /d/, /t/, /k/ and /q/) and to recognize the CV syllable. Classification experiments were performed using decision algorithms on stop consonants C and CV syllables extracted from an Arabic corpus. The results yielded to an overall stop consonants classification of 90.27% and syllables CV recognition upper than 90% for all stops.     

Proposal of new eco-manufacturing feature interaction-based methodology in CAD phase

Following the environmental crises of recent decades, a turning point in the awareness of the fragility of ecosystems has been marked, i.e., environmental awareness. This has contributed to the development of various environmental laws and regulations such as the “Waste Electrical and Electronic Equipment,” the “Restriction of Hazardous Substances,” and the “Registration, Evaluation, Authorization and Restriction of Chemicals” regulations and the “Energy Using Products” Directives. Our work contributes to the development of eco-friendly product manufacturing processes. In order to estimate and optimize the environmental impacts of a product, most of the methodologies, concepts, and tools that integrate computer-aided design (CAD) and life cycle assessment systems generally exploit the feature technology at the level of each feature independently of the others, i.e., “microplanning.” The feature interaction technology (FIT) is treated only in few studies, but it is pivotal in the eco-manufacturing process. In this paper, we propose a new manufacturing-scenarios-based methodology by using FIT and a Multi-criteria Decision Support Method (MCDSM), which helps manufacturers maintain their marketplaces by producing goods in an eco-friendly way. In fact, this methodology helps designers choose from the CAD design phase the most ecological manufacturing process from possible existent scenarios in real time.     

Formation of assimilable organic carbon (AOC) and specific natural organic matter (NOM) fractions during ozonation of phytoplankton

Ozonation of natural surface water increases the concentration of oxygen-containing low molecular weight compounds. Many of these compounds support microbiological growth and as such are termed assimilable organic carbon (AOC). Phytoplankton can contribute substantially to the organic carbon load when surface water is used as source for drinking water treatment. We have investigated dissolved organic carbon (DOC) formation from the ozonation of a pure culture of Scenedesmus vacuolatus under defined laboratory conditions, using a combination of DOC fractionation, analysis of selected organic acids, aldehydes and ketones, and an AOC bioassay. Ozonation of algae caused a substantial increase in the concentration of DOC and AOC, notably nearly instantaneously upon exposure to ozone. As a result of ozone exposure the algal cells shrunk, without disintegrating entirely, suggesting that DOC from the cell cytoplasm leaked through compromised cell membranes. We have further illustrated that the specific composition of newly formed AOC (as concentration of organic acids, aldehydes and ketones) in ozonated lake water differed in the presence and absence of additional algal biomass. It is therefore conceivable that strategies for the removal of phytoplankton before pre-ozonation should be considered during the design of drinking water treatment installations, particularly when surface water is used.     

The LiyNi0.2Mn0.2Co0.6O2 electrode materials: A structural and magnetic study

Layered LiNi_0.2Mn_0.2Co_0.6O₂ phase, belonging to a solid solution between LiNi_1/2Mn_1/2O₂ and LiCoO₂ most commercialized cathodes, was prepared via the combustion method at 900 °C for a short time (1 h). Structural and magnetic properties of this material during chemical extraction were investigated. The powders adopted the α-NaFeO₂ structure with almost none of the well-known Li/Ni cation disorder. The analysis of the magnetic properties in the paramagnetic domain agrees with the combination of Ni²⁺ (S = 1), Co³⁺ (S = 0) and Mn⁴⁺ (S = 3/2) spin-only values. X-ray analysis of the chemically delithiated Li_yNi_0.2Mn_0.2Co_0.6O₂ reveals no structural transition. The process of lithium extraction from and insertion into LiNi_0.2Mn_0.2Co_0.6O₂ was discussed on the basis of ex situ EPR experiments and magnetic susceptibility. Oxidation of Ni²⁺ (S = 1) to Ni³⁺ (S = 1/2) and to Ni⁴⁺ (S = 0) was observed upon lithium removal.     

Yttria-stabilized zirconia in-flight particle characteristics under vacuum plasma spray conditions

This paper deals with the diagnostic yttria-stabilized zirconia (YSZ) in-flight particles in Vacuum Plasma Spray (VPS) process using an optical measurement device. Particle velocity, temperature and diameter were correlated to spray distance under a fixed chamber pressure of about 14 kPa. Experiments were carried out with a two-color pyrometer. Results show that correlations can be satisfactory described with linear relationships. Particle velocity and temperature decrease when increasing spray distance whereas particle diameter exhibits a linear increase with the spray distance.     

Region-rejection based heuristics for the capacitated multi-source Weber problem

A new type of constructive and adaptive heuristics is put forward to generate initial solutions for the capacitated multi-source Weber problem. This technique is based on guiding the search by constructing restricted regions that forbid new locations to be sited too close to the previously found locations. In this work, a restricted region is represented by a circle whose radius is initially set to a fixed value, based on the sparsity of the customers and the number of facilities, and then a scheme that dynamically adjusts the radius at each facility is proposed. A discretisation technique that divides a continuous space into a discrete number of cells while embedding the use of restricted regions within the search is also presented. The experiments show that the proposed region-rejection methods, though simple and easy to understand, provide encouraging results with regard to both solution quality and computational effort. Some future research avenues are also briefly highlighted.