Showing papers by "Celal Bayar University published in 2017"

A feature selection model based on genetic rank aggregation for text sentiment classification

Abstract: Sentiment analysis is an important research direction of natural language processing, text mining and web mining which aims to extract subjective information in source materials The main challenge encountered in machine learning method-based sentiment classification is the abundant amount of data available This amount makes it difficult to train the learning algorithms in a feasible time and degrades the classification accuracy of the built model Hence, feature selection becomes an essential task in developing robust and efficient classification models whilst reducing the training time In text mining applications, individual filter-based feature selection methods have been widely utilized owing to their simplicity and relatively high performance This paper presents an ensemble approach for feature selection, which aggregates the several individual feature lists obtained by the different feature selection methods so that a more robust and efficient feature subset can be obtained In order to aggregate the individual feature lists, a genetic algorithm has been utilized Experimental evaluations indicated that the proposed aggregation model is an efficient method and it outperforms individual filter-based feature selection methods on sentiment classification

A hybrid ensemble pruning approach based on consensus clustering and multi-objective evolutionary algorithm for sentiment classification

Abstract: Sentiment analysis is a critical task of extracting subjective information from online text documents. Ensemble learning can be employed to obtain more robust classification schemes. However, most approaches in the field incorporated feature engineering to build efficient sentiment classifiers.The purpose of our research is to establish an effective sentiment classification scheme by pursuing the paradigm of ensemble pruning. Ensemble pruning is a crucial method to build classifier ensembles with high predictive accuracy and efficiency. Previous studies employed exponential search, randomized search, sequential search, ranking based pruning and clustering based pruning. However, there are tradeoffs in selecting the ensemble pruning methods. In this regard, hybrid ensemble pruning schemes can be more promising.In this study, we propose a hybrid ensemble pruning scheme based on clustering and randomized search for text sentiment classification. Furthermore, a consensus clustering scheme is presented to deal with the instability of clustering results. The classifiers of the ensemble are initially clustered into groups according to their predictive characteristics. Then, two classifiers from each cluster are selected as candidate classifiers based on their pairwise diversity. The search space of candidate classifiers is explored by the elitist Pareto-based multi-objective evolutionary algorithm.For the evaluation task, the proposed scheme is tested on twelve balanced and unbalanced benchmark text classification tasks. In addition, the proposed approach is experimentally compared with three ensemble methods (AdaBoost, Bagging and Random Subspace) and three ensemble pruning algorithms (ensemble selection from libraries of models, Bagging ensemble selection and LibD3C algorithm). Results demonstrate that the consensus clustering and the elitist pareto-based multi-objective evolutionary algorithm can be effectively used in ensemble pruning. The experimental analysis with conventional ensemble methods and pruning algorithms indicates the validity and effectiveness of the proposed scheme.

Internalized stigma in psoriasis: A multicenter study

Abstract: Internalized stigma is the adoption of negative attitudes and stereotypes of the society regarding a person's illness. It causes decreased self-esteem and life-satisfaction, increased depression and suicidality, and difficulty in coping with the illness. The primary aim of this study was to investigate the internalized stigma state of psoriatic patients and to identify the factors influencing internalized stigma. The secondary aim was to identify the correlation of internalized stigma with quality of life and perceived health status. This multicentre, cross-sectional study comprised 1485 patients. There was a significant positive correlation between mean values of Psoriasis Internalized Stigma Scale (PISS) and Psoriasis Area and Severity Index, Body Surface Area, Dermatological Life Quality Index and General Health Questionnaire-12 (P < 0.001 in all). Lower percieved health score (P = 0.001), early onset psoriasis (P = 0.016), family history of psoriasis (P = 0.0034), being illiterate (P < 0.001) and lower income level (P < 0.001) were determinants of high PISS scores. Mean PISS values were higher in erythrodermic and generalized pustular psoriasis. Involvement of scalp, face, hand, genitalia and finger nails as well as arthropathic and inverse psoriasis were also related to significantly higher PISS scores (P = 0.001). Our findings imply that psoriatic patients experience high levels of internalized stigma which are associated with psoriasis severity, involvement of visible body parts, genital area, folds or joints, poorer quality of life, negative perceptions of general health and psychological illnesses. Therefore, internalized stigma may be one of the major factors responsible from psychosocial burden of the disease.

Twitter fake account detection

Abstract: Social networking sites such as Twitter and Facebook attracts millions of users across the world and their interaction with social networking has affected their life. This popularity in social networking has led to different problems including the possibility of exposing incorrect information to their users through fake accounts which results to the spread of malicious content. This situation can result to a huge damage in the real world to the society. In our study, we present a classification method for detecting the fake accounts on Twitter. We have preprocessed our dataset using a supervised discretization technique named Entropy Minimization Discretization (EMD) on numerical features and analyzed the results of the Naive Bayes algorithm.

Species richness effects on grassland recovery from drought depend on community productivity in a multisite experiment

Abstract: Biodiversity can buffer ecosystem functioning against extreme climatic events, but few experiments have explicitly tested this. Here, we present the first multisite biodiversity × drought manipulation experiment to examine drought resistance and recovery at five temperate and Mediterranean grassland sites. Aboveground biomass production declined by 30% due to experimental drought (standardised local extremity by rainfall exclusion for 72–98 consecutive days). Species richness did not affect resistance but promoted recovery. Recovery was only positively affected by species richness in low-productive communities, with most diverse communities even showing overcompensation. This positive diversity effect could be linked to asynchrony of species responses. Our results suggest that a more context-dependent view considering the nature of the climatic disturbance as well as the productivity of the studied system will help identify under which circumstances biodiversity promotes drought resistance or recovery. Stability of biomass production can generally be expected to decrease with biodiversity loss and climate change.

Ohmic heating assisted vacuum evaporation of pomegranate juice: Electrical conductivity changes

Abstract: A novel electrical heating method, named as ohmic heating, was successfully integrated to vacuum evaporation system, and pomegranate juice was concentrated until its total soluble dry matter content reached to 40% by applying three different voltage gradients (7.5, 10, and 12.5 V/cm) at 180 mm Hg absolute pressure in this system. Total evaporation times were determined as 152, 78, and 53 min at the voltage gradients of 7.5, 10, and 12.5 V/cm, respectively. The concentration time of pomegranate juice was shortened about 56% by ohmic heating relative to conventional evaporation. The electrical conductivity values were increased up to reach 25% TSDM for 7.5 V/cm and at 35% TSDM for 10 V/cm, then showed a decreasing pattern. However, it was constant (0.55 ± 0.01 S/m) during evaporation process at 12.5 V/cm. It is recommended that the ohmic heating method could be successfully integrated to vacuum evaporation process to shorten the processing time significantly. Industrial relevance Ohmic heating has been utilized as alternative method for the purpose of heating, pasteurization, cooking etc., and relatively better products can be obtained by ohmic heating. Nowadays, energy efficient systems are needed in concentrated juice production. The integration of ohmic heating to the conventional vacuum systems could serve the production high quality juice concentrates with efficient use of energy. In present study, ohmic heating assisted vacuum evaporation of pomegranate juice was conducted, successfully. This method decreased the total process time for juice concentration, which is critically important for industrial scale productions. This novel method can be implemented to the fruit juice production lines by taking into account of design characteristics of ohmic systems such as electrical conductivity changes depending on both temperature and total soluble solids concentration. The effects of main process parameter, named as voltage gradient, on electrical conductivity changes during electrical heating assisted evaporation process has been also reported in the present study.

Ttc-3600

Abstract: Owing to the rapid growth of the World Wide Web, the number of documents that can be accessed via the Internet explosively increases with each passing day. Considering news portals in particular, sometimes documents related to categories such as technology, sports and politics seem to be in the wrong category or documents are located in a generic category called others. At this point, text categorization TC, which is generally addressed as a supervised learning task is needed. Although there are substantial number of studies conducted on TC in other languages, the number of studies conducted in Turkish is very limited owing to the lack of accessibility and usability of datasets created. In this paper, a new dataset named TTC-3600, which can be widely used in studies of TC of Turkish news and articles, is created. TTC-3600 is a well-documented dataset and its file formats are compatible with well-known text mining tools. Five widely used classifiers within the field of TC and two feature selection methods are evaluated on TTC-3600. The experimental results indicate that the best accuracy criterion value 91.03% is obtained with the combination of Random Forest classifier and attribute ranking-based feature selection method in all comparisons performed after pre-processing and feature selection steps. The publicly available TTC-3600 dataset and the experimental results of this study can be utilized in comparative experiments by other researchers.

Mixed convection in a partially heated triangular cavity filled with nanofluid having a partially flexible wall and internal heat generation

Abstract: Numerical study of mixed convection in a partially heated nanofluid-filled lid driven cavity with internal heat generation and having a partial flexible wall was performed. The bottom wall of the triangular enclosure is moving with constant speed and left vertical wall is partially heated. The inclined wall of the cavity is cooled and partially flexible. The governing equations are solved with Galerkin weighted residual finite element method. The effects of Richardson number (between 0.05 and 50), internal Rayleigh number (between 10 4 and 10 8 ), size and elastic modulus of the partial flexible wall and nanoparticle volume fraction (between 0 and 0.04) on the fluid flow and heat transfer were numerically investigated. It was observed that the local and averaged heat transfer reduce as the value of the Richardson number and internal Rayleigh number increase. As the value of the elastic modulus of the inclined wall and nanoparticle volume fraction increase, local and average heat transfer enhance. The discrepancy between the averaged Nusselt number increase for different sizes for the lower values of elastic modulus of the flexible wall. When heat transfer process is effective adding nanoparticles to the base fluid is advantageous.

Building bridges for innovation in ageing : Synergies between Action Groups of the EIP on AHA

Abstract: The Strategic Implementation Plan of the European Innovation Partnership on Active and Healthy Ageing (EIP on AHA) proposed six Action Groups. After almost three years of activity, many achievements have been obtained through commitments or collaborative work of the Action Groups. However, they have often worked in silos and, consequently, synergies between Action Groups have been proposed to strengthen the triple win of the EIP on AHA. The paper presents the methodology and current status of the Task Force on EIP on AHA synergies. Synergies are in line with the Action Groups’ new Renovated Action Plan (2016-2018) to ensure that their future objectives are coherent and fully connected. The outcomes and impact of synergies are using the Monitoring and Assessment Framework for the EIP on AHA (MAFEIP). Eight proposals for synergies have been approved by the Task Force: Five cross-cutting synergies which can be used for all current and future synergies as they consider overarching domains (appropriate polypharmacy, citizen empowerment, teaching and coaching on AHA, deployment of synergies to EU regions, Responsible Research and Innovation), and three cross-cutting synergies focussing on current Action Group activities (falls, frailty, integrated care and chronic respiratory diseases).

Fluid–structure-magnetic field interaction in a nanofluid filled lid-driven cavity with flexible side wall

Abstract: In this study MHD flow in a lid driven nanofluid filled square cavity with a flexible side wall is numerically investigated. The top wall of the cavity is colder than the bottom wall and it moves in the + x direction with constant speed. Other walls of the cavity are insulated. The finite element formulation is utilized to solve the governing equations. The Arbitrary-Lagrangian-Eulerian method is used to describe the fluid motion with the flexible wall of the cavity in the fluid–structure interaction model. The influence of the Young’s modulus of the flexible wall on the flow and heat transfer characteristics are numerically investigated for the following parameters: ( 10 4 N / m 2 ≤ E ≤ 2.5 × 10 5 N / m 2 ), with a Richardson number of ( 0.01 ≤ R i ≤ 5 ), a Hartmann number of ( 0 ≤ H a ≤ 50 ) and a volume fraction of the solid particles given by ( 0 ≤ ϕ ≤ 0.04 ). The effect of Brownian motion on the effective thermal conductivity of the nanofluid is taken into account. Averaged heat transfer decreases with increasing Hartmann number and decreasing Richardson numbers. As the Young’s modulus of the flexible wall decreases, the averaged heat transfer increases and 66.5% of the heat transfer enhancement is obtained for E = 10 4 N/m 2 compared with E = 2.5 × 10 5 N/m 2 . An averaged heat transfer enhancement of 33.87% is obtained for a solid volume fraction of 4% compared to the base fluid for the fluid–structure model coupled with the magnetic field.

Magnetic Resonance Imaging and Clinical Outcomes of Laser Therapy, Ultrasound Therapy, and Extracorporeal Shock Wave Therapy for Treatment of Plantar Fasciitis: A Randomized Controlled Trial

Abstract: We determined and compared the effectiveness of low-level laser therapy (LLLT), therapeutic ultrasound (US) therapy, and extracorporeal shock wave therapy (ESWT) using magnetic resonance imaging (MRI). We performed a randomized, prospective, comparative clinical study. A total of 60 patients with a diagnosis of chronic plantar fasciitis were divided randomly into 3 treatment groups: group 1 underwent 15 sessions of LLLT (8 J/cm 2 ; 830 nm); group 2 underwent 15 sessions of continuous US (1 mHz; 2 W/cm 2 ); and group 3 underwent 3 sessions of ESWT (2000 shocks). All patients were assessed using the visual analog scale (VAS), heel tenderness index (HTI), American Orthopaedic Foot and Ankle Society (AOFAS) ankle-hindfoot scale, Roles–Maudsley score, and MRI before and 1 month after treatment. The primary efficacy success criterion was the percentage of decrease in heel pain of >60% from baseline at 1 month after treatment for ≥2 of the 3 heel pain (VAS) measurements. Significant improvement was measured using the mean VAS, AOFAS scale, and HTI scores for all 3 groups. The thickness of the plantar fascia had decreased significantly on MRI in all 3 groups. The treatment success rate was 70.6% in the LLLT group, 65% in the ESWT group, and 23.5% in the US group. LLLT and ESWT proved significantly superior to US therapy using the primary efficacy criterion ( p = .006 and p = .012, respectively), with no significant difference between the LLLT and ESWT groups ( p > .05). The treatment of chronic plantar fasciitis with LLLT and ESWT resulted in similar outcomes and both were more successful than US therapy in pain improvement and functional outcomes.

Fatty acid composition and anticancer activity in colon carcinoma cell lines of Prunus dulcis seed oil.

Abstract: Context: Almond oil is used in traditional and complementary therapies for its numerous health benefits due to high unsaturated fatty acids content.Objectives: This study investigated the composition and in vitro anticancer activity of almond oil from Northern Cyprus and compared with almond oil from Turkey.Materials and methods: Almond oil from Northern Cyprus was obtained by supercritical CO2 extraction and analyzed by GC-MS. Almond oil of Turkey was provided from Turkish pharmacies. Different concentrations of almond oils were incubated for 24 and 48 h with Colo-320 and Colo-741 cells. Cell growth and cytotoxicity were measured by MTT assays. Anticancer and antiprolifetarive activities of almond oils were investigated by immunocytochemistry using antibodies directed against to BMP-2, β-catenin, Ki-67, LGR-5 and Jagged 1.Results: Oleic acid (77.8%; 75.3%), linoleic acid (13.5%; 15.8%), palmitic acid (7.4%; 6.3%), were determined as the major compounds of almond oil from Northern Cyprus and Turke...

Disassembly line balancing using linear physical programming

Abstract: Disassembly is the separation of a product into its constituent parts in a systematic way. It has gained importance recently due to its vital importance in product recovery. Cost-effective implementation of disassembly operation has a direct impact on the profitability of product recovery activities (recycling, remanufacturing etc.). Although it is possible to carry out disassembly operations in a disassembly station or in a disassembly cell, the highest productivity is achieved in a disassembly line. The output of a disassembly line can be maximised only if the line is balanced. A linear physical programming-based disassembly line balancing method is proposed in this study. This method was used to balance a mixed-model disassembly line and the effectiveness of the method was illustrated by analysing the results.

Electrically conductive borate-based bioactive glass scaffolds for bone tissue engineering applications.

Abstract: In this study, electrically conductive, borate-based, porous 13-93B3 bioactive glass composite scaffolds were prepared using a polymer foam replication technique For this purpose, a slurry contain

An improved ant algorithm with LDA-based representation for text document clustering

Abstract: Document clustering can be applied in document organisation and browsing, document summarisation and classification. The identification of an appropriate representation for textual documents is extremely important for the performance of clustering or classification algorithms. Textual documents suffer from the high dimensionality and irrelevancy of text features. Besides, conventional clustering algorithms suffer from several shortcomings, such as slow convergence and sensitivity to the initial value. To tackle the problems of conventional clustering algorithms, metaheuristic algorithms are frequently applied to clustering. In this paper, an improved ant clustering algorithm is presented, where two novel heuristic methods are proposed to enhance the clustering quality of ant-based clustering. In addition, the latent Dirichlet allocation LDA is used to represent textual documents in a compact and efficient way. The clustering quality of the proposed ant clustering algorithm is compared to the conventional clustering algorithms using 25 text benchmarks in terms of F-measure values. The experimental results indicate that the proposed clustering scheme outperforms the compared conventional and metaheuristic clustering methods for textual documents.