Searcharxiv⌕ Search

arXiv subjects

Sami Ullah

Publications and source records attributed to Sami Ullah.

7 recordsLinked to original sources
arXiv2026

Transforming Heart Disease Prediction with Advanced Machine Learning Techniques

Heart disease remains the leading cause of mortality globally, necessitating early and accurate detection to improve patient outcomes. This research focuses on the predictive analysis of heart disease using machine learning (ML) techniques, comparing the performance of multiple classifiers to identify the most accurate and least error-prone method. Two datasets from UCI and Kaggle repositories were utilized, each containing 14 attributes related to heart health indicators. Techniques including J48, Naive Bayes, Logistic Regression, Simple Cart, Bagging, Decision Stump, AdaBoost, Artificial Neural Networks, and Support Vector Machine (SVM) were applied. Evaluation metrics such as Mean Absolute Error (MAE), Relative Absolute Error (RAE), accuracy, precision, recall, and F-measure were used for performance comparison. Results revealed that SVM achieved the highest performance on the UCI dataset, while Simple Cart performed best on the Kaggle dataset, offering the highest accuracy and lowest error rates. The research work concludes that ML models, when properly tuned and validated, can significantly assist in the early diagnosis of heart disease, offering critical support for clinical decision-making. Future work may involve hybrid approaches and the use of more recent datasets to further improve prediction accuracy.

cs.LG↗
arXiv2024

A Fusion-Driven Approach of Attention-Based CNN-BiLSTM for Protein Family Classification -- ProFamNet

Advanced automated AI techniques allow us to classify protein sequences and discern their biological families and functions. Conventional approaches for classifying these protein families often focus on extracting N-Gram features from the sequences while overlooking crucial motif information and the interplay between motifs and neighboring amino acids. Recently, convolutional neural networks have been applied to amino acid and motif data, even with a limited dataset of well-characterized proteins, resulting in improved performance. This study presents a model for classifying protein families using the fusion of 1D-CNN, BiLSTM, and an attention mechanism, which combines spatial feature extraction, long-term dependencies, and context-aware representations. The proposed model (ProFamNet) achieved superior model efficiency with 450,953 parameters and a compact size of 1.72 MB, outperforming the state-of-the-art model with 4,578,911 parameters and a size of 17.47 MB. Further, we achieved a higher F1 score (98.30% vs. 97.67%) with more instances (271,160 vs. 55,077) in fewer training epochs (25 vs. 30).

q-bio.QM↗
arXiv2018

Extrinsic n-type semiconductor transition in ZrSe2 with the metallic character through hafnium substitution

Two dimensional layered materials exhibit versatile electronic properties in their different phases. The intrinsic electronic properties of these materials can be modulated through doping or intercalation. In this study, we investigated the electronic properties of Hf doped ZrSe2 single crystals using angle-resolved photoemission spectroscopy (ARPES) combined with first principles density functional theory (DFT) calculations. It is observed that the valence band maxima of ZrSe2, located below the Fermi level, undergo a significant change with the introduction of Hf substitution. Hf can introduce extra charges into the conduction band, rather than making a mixed structure of HfSe2 and ZrSe2 band structure, which can cross the Fermi level. Compared to the semiconducting band structure of ZrSe2, we observed that the conduction band crosses the Fermi level at the high symmetry M point in Hf-doped ZrSe2. This suggests an increase of electron type carriers around the Fermi level, resulting in an extrinsic charge carrier density in the conduction band, which can form a metallic behaviour. It can be noticed that the Hf cations can create disorder in the form of excess atoms of Zr, which yields more carriers in the conduction band in the shape of smeared bands. The tails of the smeared band occupied the d-orbitals extended into the Fermi level and left the d band below. Similarly, the electrical resistance measurements further confirm the metallic-like character of Hf doped ZrSe2 compared to the semiconductor ZrSe2, indicating increased carriers. This metallic like behavior is suggested to be predisposed by the extrinsic electrons induced by the substitutional disorder. This study further demonstrates the possibility of band gap engineering through heavy metal doping in 2D materials.

cond-mat.mtrl-sci↗
arXiv2017

Three-component bosons in TiS, ZrSe and HfTe

Topological semimetals with several types of three-dimensional (3D) fermion of electrons, such as Dirac fermions, Weyl fermions, Dirac nodal lines and triply degenerate nodal points have been theoretically predicted and then experimentally discovered in the electronic structures of a series of solid crystals. In analogy of various typical fermions, topological mechanical states with two type of bosons, Dirac and Weyl bosons, were also experimentally reported in some macroscopic systems of kHz frequency and with a type of doubly-Weyl phonons in atomic vibrational framework of THz frequency of solid crystal was also recently predicted. However, to date no triply degenerate nodal point of phonon beyond the conventional Dirac, Weyl and doubly-Weyl phonons has been reported. Here, through first-principles calculations, we have reported on the prediction that the WC-type TiS, ZrSe, and HfTe commonly host the unique triply degenerate nodal point of phonon in THz frequency due to the occurrence of the phonon band inversion between the doubly degenerate planar vibrational mode and the singlet vertical vibrational mode at the boundary A point of the bulk Brillouin zone. Quasiparticle excitations near this triply degenerate nodal point of phonons are three-component bosons, different from the known classifications. The underlying mechanism can be attributed to the leading role of the comparable atomic masses of constituent elements in compounds in competition with the interatomic interaction. Additionally, the electronic structures in their bulk crystals exhibit the coexisted triply degenerate nodal point and Weyl fermions. The novel coexistence of three-component bosons, three-component fermions and Weyl fermions in these materials thus suggest an enriched platform for studying the interplay between different types of fermions and bosons.

cond-mat.mtrl-sci↗
arXiv2017

Topological nodal line states and a potential catalyst of hydrogen evolution in the TiSi family

Topological nodal line (DNL) semimetals, formed by a closed loop of the inverted bands in the bulk, result in the nearly flat drumhead-like surface states with a high electronic density near the Fermi level. The high catalytic active sites associated with the high electronic densities, the good carrier mobility, and the proper thermodynamic stabilities with $\Delta G_{H^*}$$\approx$0 are currently the prerequisites to seek the alternative candidates to precious platinum for catalyzing electrochemical hydrogen (HER) production from water. Within this context, it is natural to consider whether or not the DNLs are a good candidate for the HER because its non-trivial surface states provide a robust platform to activate possibly chemical reactions. Here, through first-principles calculations we reported on a new DNL TiSi-type family with a closed Dirac nodal line consisting of the linear band crossings in the $k_y$ = 0 plane. The hydrogen adsorption on the (010) and (110) surfaces yields the $\Delta G_{H^*}$ to be almost zero. The topological charge carries have been revealed to participate in this HER. The results are highlighting that TiSi not only is a promising catalyst for the HER but also paves a new routine to design topological quantum catalyst utilizing the topological DNL-induced surface bands as active sites, rather than edge sites-, vacancy-, dopant-, strain-, or heterostructure-created active sites.

cond-mat.mtrl-sci↗
arXiv2016

Weyl Ferroelectric Semimetal

The recent discoveries of ferroelectric metal and Weyl semimetal (WSM) have stimulated a natural question: whether these two exotic states of matter can coexist in a single material or not. These two discoveries ensure us that physically it is possible since both of them share the same necessary condition, the broken inversion symmetry. Here, by using first-principles calculations, we demonstrate that the experimentally synthesized nonmagnetic HgPbO$_3$ represents a unique example of such hybrid "\emph{Weyl ferroelectric semimetal}". Its centrosymmetric $R\bar{3}c$ phase will undergo a ferroelectric phase transition to the ferroelectric $R3c$ structure. Both phases are metallic and the ferroelectric phase owns a spontaneous polarization of 33 $\mu$C/cm$^2$. Most importantly, it also harbors six pairs of chiral Weyl nodes around the Fermi level to be an oxide WSM. The structural symmetry broken phase transition induces a topological phase transition. The coexistence of ferroelectricity and Weyl nodes in HgPbO$_3$ is an ideal platform for exploring multiphase interaction and mutual control. The Weyl nodes can be tuned by external pulse electric field, which is promising for potential applications of integrated topotronic and ferroelectric devices.

cond-mat.mtrl-sci↗