Python Code Research Articles

MetaNet: Interpretable unknown mobile malware identification with a novel meta-features mining algorithm

The continuous emergence of malware has threatened to the Android platform and user privacy. With the evolution of the Android system and malware, it is challenging to design a method that can accurately identify the categories of sophisticated malware, including known and unknown families, as well as their obfuscated variants, given that they may be newly emerging and lack available detection knowledge. Although some methods try to use anomaly detection and zero-shot technology to identify unseen applications, they are limited to binary classification or lack the robustness, stability, universality, and interpretability in multi-class identification. To this end, we first propose a generic meta-features mining algorithm, which can discover the potential relationships between samples belonging to the same category. Then we present metaNet, a novel method leveraging meta-features to identify sophisticated Android malware. Specifically, metaNet is mainly powered by four components: (i) mExtractor is a feature collector to obtain the static and dynamic features. (ii) mProcessor is taking unique meta-features of each category from extracted features. (iii) mLearner is a machine learning suite that leverages features and meta-features to design and train a classifier called HSU-Net. (iv) mEnforcer is a flexible deployer that identifies categories of malware families in the real world. We implement a prototype of metaNet with 15K lines of Python code and compare it with state-of-the-art (SOTA) methods. The results show that it can not only achieve superior performance in terms of known families (99.52% of accuracy) and unknown families (99.31% of accuracy trained with 80% known families) for binary classification, but also perform well in multi-class identification, i.e., 99.05% and 93.45% of accuracy for known and unknown families, respectively. Furthermore, we deploy and evaluate metaNet in the real world. It can identify applications over an acceptable time and memory overheads, i.e., average of 11.8 s and 56 MB per sample with a size of 8 MB. Also, the few-shot detection and feature perturbation experiments reflect its robustness and stability benefiting from meta-features. Finally, we collect the traffic of 112 decentralized applications (DApps) belonging to 16 categories, such as finance and health, and evaluated metaNet in DApp identification. The results illustrate its applicability across various tasks. That is, it can accurately classify 94.6% and 81.36% of DApp flows in all-known and 80%-known DApp scenarios, respectively, outperforming the SOTA methods.

Python Programming for Degree Heating Weeks Estimation using Sea Surface Temperature Data from The Google Earth Engine Dataset as Coral Bleaching Analysis Tools

Abstract This study aims to develop a program to calculate Degree Heating Weeks (DHW) using Python programming and Sea Surface Temperature (SST) data available in Google Earth Engine (GEE) datasets, e.g., HyCom, NOAA OISST v2.1, MODIS-Aqua, and NOAA Pathfinder v5.3. This study contributes to providing a DHW calculator currently unavailable on the Internet to analyze coral bleaching. This program is called PyDHW. The DHW was calculated by inputting the desired location in a polygon, a specific time range, and the SST data. The results show that Python code can extract SST from the GEE dataset according to the user’s input. This SST data is the average of the SST inside the polygon area. This program calculates and shows a graph of the DHW showing coral bleaching alert levels. All these processes were performed quickly in one run. HyCom and NOAA OISST v2.1 have a long and continuous data range. NOAA OISST v2.1 is still updated in the GEE dataset rather than the others. The MODIS-Aqua contains blank time-series data for several measurements. The NOAA Pathfinder v5.3 data shows extreme change in time series data and low temperatures with different patterns from the other SST data. However, this program is still under development and needs improvements. This program is expected to help users concerned with coral research and monitoring.

GraphPyRec: A novel graph-based approach for fine-grained Python code recommendation

Artificial intelligence has been widely applied in software engineering areas such as code recommendation. Significant progress has been made in code recommendation for static languages in recent years, but it remains challenging for dynamic languages like Python as accurately determining data flows before runtime is difficult. This limitation hinders data flow analysis, affecting the performance of code recommendation methods that rely on code analysis. In this study, a graph-based Python recommendation approach (GraphPyRec) is proposed by converting source code into a graph representation that captures both semantic and dynamic information. Nodes represent semantic information, with unique rules defined for various code statements. Edges depict control flow and data flow, utilizing a child-sibling-like process and a dedicated algorithm for data transfer extraction. Alongside the graph, a bag of words is created to include essential names, and a pre-trained BERT model transforms it into vectors. These vectors are integrated into a Gated Graph Neural Network (GGNN) process of the code recommendation model, enhancing its effectiveness and accuracy. To validate the proposed method, we crawled over a million lines of code from GitHub. Experimental results show that GraphPyRec outperforms existing mainstream Python code recommendation methods, achieving Top-1, 5, and 10 accuracy rates of 68.52%, 88.92%, and 94.05%, respectively, along with a Mean Reciprocal Rank (MRR) of 0.772.

Toughening of intrinsically brittle materials by inserting arrays of voids

The aim of the current paper is to find optimum arrangements of voids that are able to catch and trap all possible cracks that might originate from a free surface. The idea is to investigate whether it is possible to apply this crack trapping mechanism to enhance the fracture toughness of intrinsically brittle materials. In doing so, the void volume shall be as small as possible. The investigation applies the crack trajectory interpolation (CTI) method, which is a new, computationally efficient approach to approximate crack trajectories near a single void or stiff particle. Under the assumption of linear superposition, the CTI-method is extended to predict the crack trajectories near arrangements of voids. This is done for uniaxial and biaxial loading conditions. The investigation shows that a symmetrical arrangement of voids should be avoided, since cracks near the symmetry line can bypass the array. Staggered void arrangements provide better conditions for crack trapping. By using elliptical voids with an aspect ratio of 2 instead of circular voids, it is possible to achieve a higher crack trapping efficiency and a lower void volume. A damage-based approach is used to determine the improvements in fracture toughness due to crack trapping for various void arrangements. The Python code of the CTI method is presented in the Appendix of this work.

Benchmarking Llama 3 70B for Code Generation: A Comprehensive Evaluation

This study benchmarks the capabilities of Llama 3 70B, a 70-billion parameter large language model (LLM), for code generation tasks. To effectively train and fine-tune this massive model, we integrate PyTorch Fully Sharded Data Parallel (FSDP) [1], [2] for distributed training and Quantized Low-Rank Adaptation (Q-LoRA) [7] for efficient fine-tuning. We address challenges associated with distributed training, including communication overhead and synchronization complexities, through optimization strategies like gradient accumulation, optimizer state sharding, and mixed precision training. Additionally, we employ advanced training techniques such as Curriculum Learning, Dynamic Batch Sizing, and Adaptive Optimization Algorithms to enhance model performance and training efficiency. Our primary focus is evaluating the performance of the fine-tuned Llama 3 70B model on two widely-recognized code generation benchmarks: HumanEval [8] and MBPP [9]. HumanEval assesses the model's ability to translate natural language problem descriptions into functionally correct code, while MBPP evaluates its proficiency in solving complex programming problems by generating accurate Python code. We present detailed performance results on these benchmarks, analyzing the model's strengths and limitations in various code generation scenarios. Furthermore, we compare the impact of our training and fine-tuning methodologies on scalability, memory efficiency, and training speed, demonstrating the feasibility and efficiency of our approach. This benchmark study offers valuable insights for researchers and practitioners exploring the application of LLMs for code generation. It provides a comprehensive evaluation of Llama 3 70B's capabilities, sheds light on the effectiveness of various training and fine-tuning techniques, and emphasizes the importance of rigorous benchmark evaluation in driving progress within this rapidly evolving field.

The Crosswise Model for Surveys on Sensitive Topics: A General Framework for Item Selection and Statistical Analysis

When surveys contain direct questions about sensitive topics, participants may not provide their true answers. Indirect question techniques incentivize truthful answers by concealing participants’ responses in various ways. The Crosswise Model aims to do this by pairing a sensitive target item with a non-sensitive baseline item, and only asking participants to indicate whether their responses to the two items are the same or different. Selection of the baseline item is crucial to guarantee participants’ perceived and actual privacy and to enable reliable estimates of the sensitive trait. This research makes the following contributions. First, it describes an integrated methodology to select the baseline item, based on conceptual and statistical considerations. The resulting methodology distinguishes four statistical models. Second, it proposes novel Bayesian estimation methods to implement these models. Third, it shows that the new models introduced here improve efficiency over common applications of the Crosswise Model and may relax the required statistical assumptions. These three contributions facilitate applying the methodology in a variety of settings. An empirical application on attitudes toward LGBT issues shows the potential of the Crosswise Model. An interactive app, Python and MATLAB codes support broader adoption of the model.

Graph neural networks for CO[formula omitted] solubility predictions in Deep Eutectic Solvents

Deep Eutectic Solvents (DESs) are a promising class of solvents for CO2 capture. DESs are complex mixtures that can be designed to optimize CO2 solubility and overall capture process efficiency. However, the vast design landscape of DES mixtures makes experimental investigation prohibitive; as such, there is a need for computational models that can quickly and efficiently navigate the design space and inform data collection efforts. In this work, we propose Graph Neural Network (GNN) models for predicting CO2 solubility for DESs; the GNN leverages a mixture graph representation that captures the molecular structure of the DES components as well as their intermolecular interactions. We compare the GNN framework against alternative architectures (neural networks, graph convolution networks, and random forests) and data representations (molecular fingerprints, sigma profiles, and graphs). We show that the proposed approach offers superior predictive performance; specifically, we show that solubility can be predicted reliably directly from molecular structure (without the need of using sigma profiles as proposed in previous studies). This result is important, as obtaining sigma profiles requires expensive density functional theory computations. We also explored the ability of GNNs to predict solubility for new DES mixtures and operating conditions. We found that the model extrapolates across temperature reliably. However, we also found deficiencies in the ability of the model to predict solubility for DES mixtures, pressures, and molar ratio not included in the training sets; we show that this is due to an inherent lack of chemical diversity in datasets available in the literature. The proposed computational capabilities can thus help navigate the design space of DES and inform data collection efforts. Our models, data, and benchmarks are shared as Python code implemented in Jupyter notebooks.

A tutorial on solving single‐leader‐multi‐follower problems using SOS1 reformulations

AbstractIn this tutorial, we consider single‐leader‐multi‐follower games in which the models of the lower‐level players have polyhedral feasible sets and convex objective functions. This situation allows for classic Karush–Kuhn–Tucker reformulations of the separate lower‐level problems, which lead to challenging single‐level reformulations of Mathematical Programing with Complementarity Constraints (MPCC) type. The main contribution of this tutorial is to present a ready‐to‐use reformulation of this MPCC using special‐ordered‐sets of type 1 (SOS1) conditions. These conditions are readily available in all modern mixed‐integer linear optimization solvers that solve the single‐leader‐multi‐follower problem to optimality. After formally stating the problem class under consideration as well as deriving its reformulations, we present explicit Python code that shows how these techniques can be realized using the solver Gurobi. Finally, we also show the effect of the SOS1‐based reformulation using the real‐world example of industrial eco‐park modeling.

Data repairing and resolution enhancement using data-driven modal decomposition and deep learning

This paper introduces a new series of methods which combine modal decomposition algorithms, such as singular value decomposition and high-order singular value decomposition, and deep learning architectures to repair, enhance, and increase the quality and precision of numerical and experimental data. A combination of two- and three-dimensional, numerical and experimental datasets are used to demonstrate the reconstruction capacity of the presented methods, showing that these methods can be used to reconstruct any type of dataset, showing outstanding results when applied to highly complex data, which is noisy. The combination of benefits of these techniques results in a series of data-driven methods which are capable of repairing and/or enhancing the resolution of a dataset by identifying the underlying physics that define the data, which is incomplete or under-resolved, filtering any existing noise. These methods and the Python codes are included in the first release of ModelFLOWs-app.11The website of the software is available at https://modelflows.github.io/modelflowsapp/.

A multimodal Transformer Network for protein-small molecule interactions enhances predictions of kinase inhibition and enzyme-substrate relationships.

The activities of most enzymes and drugs depend on interactions between proteins and small molecules. Accurate prediction of these interactions could greatly accelerate pharmaceutical and biotechnological research. Current machine learning models designed for this task have a limited ability to generalize beyond the proteins used for training. This limitation is likely due to a lack of information exchange between the protein and the small molecule during the generation of the required numerical representations. Here, we introduce ProSmith, a machine learning framework that employs a multimodal Transformer Network to simultaneously process protein amino acid sequences and small molecule strings in the same input. This approach facilitates the exchange of all relevant information between the two molecule types during the computation of their numerical representations, allowing the model to account for their structural and functional interactions. Our final model combines gradient boosting predictions based on the resulting multimodal Transformer Network with independent predictions based on separate deep learning representations of the proteins and small molecules. The resulting predictions outperform recently published state-of-the-art models for predicting protein-small molecule interactions across three diverse tasks: predicting kinase inhibitions; inferring potential substrates for enzymes; and predicting Michaelis constants KM. The Python code provided can be used to easily implement and improve machine learning predictions involving arbitrary protein-small molecule interactions.

Some results on a nonlinear fractional equation with nonlocal boundary condition

The aim of this paper is to derive sufficient conditions for the existence, uniqueness, and Hyers–Ulam stability of solutions to a new nonlinear fractional integro‐differential equation with functional boundary conditions, using several fixed‐point theorems, the multivariate Mittag‐Leffler function and Babenko's approach. A few examples are also presented to illustrate the applications of our results based on approximate values of a couple of Mittag‐Leffler functions calculated by Python codes. Furthermore, the approaches used have a wide range of applications to various fractional differential equations with initial or boundary conditions or integral equations in complete spaces.

Identifying 124 new anti-HIV drug candidates in a 37 billion-compound database: An integrated approach of machine learning (QSAR), molecular docking, and molecular dynamics simulation

Recent data from the World Health Organization reveals that in 2023, 38.8 million people were living with HIV. Within this population, there were 1.5 million new cases and 650 thousand deaths attributed to the disease. This study employs an integrated approach involving QSAR-based machine learning models, molecular docking, and molecular dynamics simulations to identify potential compounds for inhibiting the bioactivity of the CC chemokine receptor type 5 (CCR5) protein, a key entry point for the HIV virus. Using non-redundant experimental data from the CHEMBL database, 40 different machine learning algorithms were trained and the top four models (XGBoost, Histogram based gradient Boosting, Light Gradient Boosted Machine, and Extra Trees Regression) were utilized to predict anti-HIV bioactivity for 37 billion compounds in the ZINC-22 database. The screening resulted in the identification of 124 new anti-HIV drug candidates, confirmed through molecular docking and dynamics simulations. The study underscores the therapeutic potential of these compounds, paving the way for further in vitro and in vivo investigations. The convergence of machine learning and experimental findings presents a promising avenue for significant advancements in pharmaceutical research, particularly in the treatment of viral diseases such as HIV. To guarantee the reproducibility of our study, we have made the Python code (google colab) and the associated database available on GitHub. You can access them through the following link: GitHub Link: https://github.com/AlexandreCOBRE/code.

Reducing the Constrained Multibody Dynamics Problem to the Solution of a System of Ordinary Differential Equations Via Velocity Partitioning and Lie Group Integration

Abstract In multibody dynamics, formulating the equations of motion in absolute Cartesian coordinates results in a set of index-3 differential algebraic equations (DAEs). In this work, we present an approach that bypasses the DAE problem by partitioning the velocities in the system into dependent and independent coordinates, thereby reducing the task of producing the time evolution of the mechanical system to one of solving a set of ordinary differential equations (ODEs). In this approach, the independent coordinates are integrated directly, while the dependent coordinates are recovered through the kinematic constraint equations at the position and velocity levels. Notably, Lie group integration is employed to directly obtain the orientation matrix A at each time-step of the simulation. This eliminates the need to choose generalized coordinates to capture the orientation of a body, as the matrix A is a by-product of the solution algorithm. Herein, we outline the new approach and demonstrate it in conjunction with four mechanisms: a single pendulum, a double pendulum, a four-link mechanism, and a slider crank. We report on the convergence order behavior of the proposed method and compare its performance with an established method that combines coordinate partitioning with an Euler parameter formulation. The Python code developed to generate the reported results is open-source and available in a public repository for reproducibility studies.

The PDI model system for parameterizing soil hydraulic properties

AbstractThe Peters–Durner–Iden (PDI) model system for describing soil hydraulic properties (SHP) has been developed over a decade. Inspired by Rien van Genuchten's seminal work, the PDI system focuses on an efficient and simple parameterization of water retention curves and hydraulic conductivity curves (HCC) across the entire soil moisture spectrum. By combining capillary and non‐capillary components for water retention and conductivity, it aims to reconcile mathematical simplicity and insights on water adsorption and film flow in soils. Recent developments have reduced the number of free parameters of the conductivity model to zero, enhancing the model's applicability in cases of limited data availability. The first reduction was achieved by a prediction of absolute non‐capillary conductivity based on the consideration of film and corner flow on the pore scale, and the second by a prediction of absolute capillary conductivity by a capillary bundle model. This allows a complete characterization of SHP over the entire moisture range with only four retention curve parameters. The inclusion of a maximum pore size in the capillary conductivity model prevents an unrealistic drop of the HCC near saturation. This paper provides a comprehensive overview of the PDI model system, emphasizing its conceptual features and mathematical details. An Excel sheet and a Python code stored in a repository are provided for accessibility.

Equivalent sentiment measures for cross-language analysis of corporate communications

This paper presents a technique for sentiment measurement in many languages. The method allows researchers to efficiently analyze corporate documents, management reports, and financial statements using python. When the texts are written in many languages, the method extracts equivalent cross-linguistic sentiment features that can be used for statistical analysis or machine learning. We use Open Multilingual WordNet, a large lexicon organizing words into semantic groups, as the knowledge base about word equivalence in more than 200 languages. We experiment with a parallel English-French corpus and find that our senitment measures across the two languages are comparable. The method produces a consistent classification of positive and negative texts in two languages, and sentiment measure values correlate. The paper provides a detailed account of the method and python code, So that it can be applied to other languages, text mining, quantitative communication studies, and management research.•Method to create equivalent sentiment measures in multiple languages•Based on established lexicons and WordNet•Validated for English and French

SEISMIC, a Python-based code of the quantas package to calculate the phase and group acoustic velocities in crystals

The present work reports the theoretical background and the capabilities of SEISMIC, a Python code specifically developed to calculate the propagation of the sound waves inside crystalline materials. SEISMIC is a tool integrated in the Quantas package and provides a series of useful information for engineers and geophysicists, such as the phase and group velocities, the power flow angle, the enhancement factors, and the polarization vectors, using as input the elastic moduli and the density of the material. Numerical treatments of the derivatives were avoided, using analytical methods to obtain numerically stable results. The code relies only on Python numerical and graphical libraries to ensure a full cross-platform usability.

Innovative PNA-LB mediated allele-specific LAMP for KRAS mutation profiling on a compact lab-on-a-disc device

Tailored healthcare, an approach focused on individual patients, requires integrating emerging interdisciplinary technologies to develop accurate and user-friendly diagnostic tools. KRAS mutations, prevalent in various common cancers, are crucial determinants in selecting patients for novel KRAS inhibitor therapies. This study presents a novel state-of-the-art Lab-on-a-Disc system utilizing peptide nucleic acids-loop backward (PNA-LB) mediated allele-specific loop-mediated isothermal amplification (LAMP) for detecting the frequent G12D KRAS mutation, signifying its superiority over alternative mutation detection approaches. The designed Lab-on-a-Disc system demonstrated exceptional preclinical and technical precision, accuracy, and versatility. By applying varying cutoff values to PNA- LB LAMP reactions, the assay's sensitivity and specificity were increased by 80 % and 90 %, respectively. The device's key advantages include a robust microfluidic Lab-on-a-Disc design, precise rotary control, and a cutting-edge induction heating module. These features enable multiplexing of LAMP reactions with high reproducibility and repeatability, with CV% values less than 3.5 % and 5.5 %, respectively. The device offers several methods for accurate endpoint result detection, including naked-eye observation, RGB image analysis using Python code, and time of fluorescence (Tf) values. Preclinical specificity and sensitivity, assessed using different cutoffs for Eva-Green fluorescence Tf values and pH-sensitive dyes, demonstrated comparable performance to the best standard methods. Overall, this study represents a significant step towards tailoring treatment strategies for cancer patients through precise and efficient mutation detection technologies.

M-band wavelet network for machine anomaly detection from a frequency perspective

The autoencoder (AE) is widely utilized in deep anomaly detection, but it lacks explainability due to the complexity of nonlinear mapping. One approach to address this issue is incorporating wavelet theory, which shares similarities in decomposition and reconstruction procedures. However, the perfect reconstruction property of wavelet theory conflicts with AE-based anomaly detection. To tackle this problem, we introduce a novel deep anomaly detection method from a frequency perspective. A learnable M-band wavelet network (MWNet) is designed to offer a flexible frequency band structure for signal representation. Subsequently, with the aid of sparsity constraint, MWNet dynamically focuses on key components within each frequency band. A learnable hard threshold function with a threshold maximization constraint is proposed to retain the essential frequency band of normal signals. After training, the MWNet is exclusively capable of well reconstructing normal signals, thereby producing a noticeable reconstruction error difference between normal and abnormal signals. Extensive experiments on both simulated and experimental datasets validate the effectiveness of the proposed method. The corresponding Python codes are available at https://github.com/albertszg/MbandWaveletNet.

Evaluating the Quality of Serial EM Sections with Deep Learning.

Automated image acquisition can significantly improve the throughput of serial section scanning electron microscopy (ssSEM). However, image quality can vary from image to image depending on autofocusing and beam stigmation. Automatically evaluating the quality of images is, therefore, important for efficiently generating high-quality serial section scanning electron microscopy (ssSEM) datasets. We tested several convolutional neural networks for their ability to reproduce user-generated evaluations of ssSEM image quality. We found that a modification of ResNet-50 that we term quality evaluation Network (QEN) reliably predicts user-generated quality scores. Running QEN in parallel to ssSEM image acquisition therefore allows users to quickly identify imaging problems and flag images for retaking. We have publicly shared the Python code for evaluating images with QEN, the code for training QEN, and the training dataset.

Python code for modeling ARIMA-LSTM architecture with random forest algorithm

Over conventional statistical models, machine learning mechanisms are establishing themselves as a potential area for modeling and forecasting complex time series. Because it can integrate several forecasting methodologies’ capabilities, hybrid time series models are fundamental in data science. Here, we present a Python script that builds a combined architecture of the ARIMA-LSTM model with random forest technique to generate a high-accuracy prediction. This script is a step-by-step process to create a statistical and then machine learning model through statistical assumption.