The complexity and variety of language included in policy and academic documents make the automatic classification of research papers based on the United Nations Sustainable Development Goals (SDGs) somewhat difficult. Using both pre-trained and contextual word embeddings to increase semantic understanding, this study presents a complete deep learning pipeline combining Bidirectional Long Short-Term Memory (BiLSTM) and Convolutional Neural Network (CNN) architectures which aims primarily to improve the comprehensibility and accuracy of SDG text classification, thereby enabling more effective policy monitoring and research evaluation. Successful document representation via Global Vector (GloVe), Bidirectional Encoder Representations from Transformers (BERT), and FastText embeddings follows our approach, which comprises exhaustive preprocessing operations including stemming, stopword deletion, and ways to address class imbalance. Training and evaluation of the hybrid BiLSTM-CNN model on several benchmark datasets, including SDG-labeled corpora and relevant external datasets like GoEmotion and Ohsumed, help provide a complete assessment of the model’s generalizability. Moreover, this study utilizes zero-shot prompt-based categorization using GPT-3.5/4 and Flan-T5, thereby providing a comprehensive benchmark against current approaches and doing comparative tests using leading models such as Robustly Optimized BERT Pretraining Approach (RoBERTa) and Decoding-enhanced BERT with Disentangled Attention (DeBERTa). Experimental results show that the proposed hybrid model achieves competitive performance due to contextual embeddings, which greatly improve classification accuracy. The study explains model decision processes and improves openness using interpretability techniques, including SHapley Additive exPlanations (SHAP) analysis and attention visualization. These results emphasize the need to incorporate rapid engineering techniques alongside deep learning architectures for effective and interpretable SDG text categorization. With possible effects on more general uses in policy analysis and scientific literature mining, this work offers a scalable and transparent solution for automating the evaluation of SDG research.
The paper attempts to find out the elements of picaresque novel in selected English and Iraqi novels. It studies these elements in Henry Fielding’s Joseph Andrews and Adil Abduljabbar’s Arzal Hamad Al-Salim. The paper is divided into four sections. The first is an introduction to the picaresque novel. It gives a definition, a historical background, and the elements of the genre. The second section studies Fielding’s novel focusing on the elements of this type of novel and how it affects the story itself. The paper follows the novel from the beginning to the end showing these elements. The third is dedicated to Abduljabbar’s novel and how the elements of picaresque genre appear in the novel and play an important role in its developm
... Show MoreThis study investigates the impact of spatial resolution enhancement on supervised classification accuracy using Landsat 9 satellite imagery, achieved through pan-sharpening techniques leveraging Sentinel-2 data. Various methods were employed to synthesize a panchromatic (PAN) band from Sentinel-2 data, including dimension reduction algorithms and weighted averages based on correlation coefficients and standard deviation. Three pan-sharpening algorithms (Gram-Schmidt, Principal Components Analysis, Nearest Neighbour Diffusion) were employed, and their efficacy was assessed using seven fidelity criteria. Classification tasks were performed utilizing Support Vector Machine and Maximum Likelihood algorithms. Results reveal that specifi
... Show MoreSemiconductor quantum dots (QDs) have attracted tremendous attentions for their unique characteristics for solid-state lighting and thin-film display applications. A simple chemical method was used to synthesis quantum dots (QDs) of zinc sulfide (ZnS) with low cost. The XRD) shows cubic phase of the prepared ZnS with an average particles size of (3-29) nm. In UV-Vis. spectra observed a large blue shift over 38 nm. The band gaps energy (Eg) was 3.8 eV and 3.37eV from the absorption and photoluminescence (PL) respectively which larger than the Eg for bulk. QDs-LED hybrid devices were fabricated using ITO/ PEDOT: PSS/ Poly-TPD/ ZnS-QDs/ with different electron transport layers and cathode of LiF/Al layers. The EL spectrum reveals a bro
... Show MoreEnsuring security, integrity, and reliability of the election process consider as the main challenges in the electronic voting system. This paper describes the e-voting system by integrating the biometric authentication, advanced encryption, and watermarking techniques towards meeting such challenges. The system employs the fingerprint authentication by utilizing the Scale-Invariant Feature Transform (SIFT) for verifying the identity of the voter to ensure genuineness and non-repudiation of the service. The vote will be encrypted with the AES-GCM technique to be employed in securing the voting process, thus ensuring both data privacy and integrity. Hybrid Blind Watermarking employs the technique of Discrete Wavelet Transform (DWT) a
... Show MoreIn this research a study of the effect of quality, sequential and directional layers for three types of fibers are:(Kevlar fibers-49 woven roving and E- glass fiber woven roving and random) on the fatigue property using epoxy as matrix. The test specimens were prepared by hand lay-up method the epoxy resin used as a matrix type (Quick mast 105) in prepared material composit . Sinusoidal wave which is formed of variable stress amplitudes at 15 Hz cycles was employed in the fatigue test ( 10 mm )and (15mm) value 0f deflection arrival to numbers of cycle failure limit, by rotary bending method by ( S-N) curves this curves has been determined ( life , limit and fa
... Show MoreThis paper proposes two hybrid feature subset selection approaches based on the combination (union or intersection) of both supervised and unsupervised filter approaches before using a wrapper, aiming to obtain low-dimensional features with high accuracy and interpretability and low time consumption. Experiments with the proposed hybrid approaches have been conducted on seven high-dimensional feature datasets. The classifiers adopted are support vector machine (SVM), linear discriminant analysis (LDA), and K-nearest neighbour (KNN). Experimental results have demonstrated the advantages and usefulness of the proposed methods in feature subset selection in high-dimensional space in terms of the number of selected features and time spe
... Show MoreImage compression is one of the data compression types applied to digital images in order to reduce their high cost for storage and/or transmission. Image compression algorithms may take the benefit of visual sensitivity and statistical properties of image data to deliver superior results in comparison with generic data compression schemes, which are used for other digital data. In the first approach, the input image is divided into blocks, each of which is 16 x 16, 32 x 32, or 64 x 64 pixels. The blocks are converted first into a string; then, encoded by using a lossless and dictionary-based algorithm known as arithmetic coding. The more occurrence of the pixels values is codded in few bits compare with pixel values of less occurre
... Show MoreAddition of bioactive materials such as Titanium oxide (TiO2), and incorporation of bio inert ceramic such as alumina (Al2O3), into polyetheretherketone (PEEK) has been adopted as an effective approach to improve bone-implant interfaces. In this paper, hot pressing technique has been adopted as a production method. This technique gave a homogenous distribution of the additive materials in the proposed composite biomaterial. Different compositions and compounding temperatures have been applied to all samples. Mechanical properties and animal model have been studied in all different production conditions. The results of these new TiO2/Al2O3/PEEK biocomposites with different
... Show More