检索结果-内蒙古大学图书馆

2021 International Conference on data analytics for Business and Industry, ICDABI 2021

作者： Mohammed, Abdulrahim Khedr, Ahmed Alhaj, Duaa Al Khalifa, Reem Alqaddoumi, Abdulla University of Bahrain Big Data Science and Analytics College of Science Sakhir Bahrain University of Bahrain College of Information Technology Department of Computer Science Sakhir Bahrain

ISBN: (纸本)9781665416566

Time-series data has a natural chronological arrangement with modeling and cross-validation techniques, highly dependent on sequential processing of data, which challenges its parallelization. Since the running time of the modeling algorithm largely consists of a main operation of solving the modeling algorithm several times with various training and testing dataset sizes, it is the prime target for optimizing the running time. This research presents a parallel implementation of time-series cross-validation on a rolling basis that involves using a subset of the dataset for training purposes, expanding each dataset with each run to obtain test accuracy. The evaluation is conducted using parallel speedup, parallel efficiency, and computational time. The proposed algorithm distributes the task of modeling among several processors to run independently and obtain results. The computational time recorded ranged from 15.47 seconds at best with five processors to 28.33 seconds at worst when done with a single processor. The parallel system speedup was found to be sub-linear with an efficiency decreasing with the increase in the number of processors. © 2021 IEEE.

关键词： Time series

来源：评论

学校读者我要写书评

暂无评论

Parallel Branch-and-Bound-Comparing Serial and OpenMP Implementations in a Color Sorting Game

Parallel Branch-and-Bound-Comparing Serial and OpenMP Implem...

引用

2021 International Conference on data analytics for Business and Industry, ICDABI 2021

作者： Khalid, Aisha Abdulrahman, Hajar Khedr, Ahmed Alqaddoumi, Abdulla University of Bahrain Big Data Science and Analytics College of Science Sakhir Bahrain University of Bahrain College of Information Technology Department of Computer Science Sakhir Bahrain

ISBN: (纸本)9781665416566

Branch-and-Bound algorithm is very diverse in its applications, one of which is the gaming industry, especially for complicated games that need optimization in their search for a solution from a large search space. For a thorough study, the Branch-and-Bound algorithm will be implemented in a serial and parallel (using Open Multi-Processing) approach to compare time required to solve the game. Results showed that for easy levels, the serial program performed better, while for difficult levels (with more flasks and colors) the parallel program outperformed the serial program with a speed up of more than 1,000x. Future work can replicate the algorithm on different types of games and explore different approaches such as Backtracking in a parallel approach. © 2021 IEEE.

关键词： Branch and bound method

来源：评论

学校读者我要写书评

暂无评论

PipeMoE: Accelerating Mixture-of-Experts through Adaptive Pipelining

PipeMoE: Accelerating Mixture-of-Experts through Adaptive Pi...

引用

IEEE Annual Joint Conference: INFOCOM, IEEE computer and Communications Societies

作者： Shaohuai Shi Xinglin Pan Xiaowen Chu Bo Li School of Computer Science and Technology Harbin Institute of Technology Shenzhen Department of Computer Science Hong Kong Baptist University Data Science and Analytics Thrust The Hong Kong University of Science and Technology (Guangzhou) Department of Computer Science and Engineering The Hong Kong University of Science and Technology

Large models have attracted much attention in the AI area. The sparsely activated mixture-of-experts (MoE) technique pushes the model size to a trillion-level with a sub-linear increase of computations as an MoE layer can be equipped with many separate experts, but only one or two experts need to be trained for each input data. However, the feature of dynamically activating experts of MoE introduces extensive communications in distributed training. In this work, we propose PipeMoE to adaptively pipeline the communications and computations in MoE to maximally hide the communication time. Specifically, we first identify the root reason why a higher pipeline degree does not always achieve better performance in training MoE models. Then we formulate an optimization problem that aims to minimize the training iteration time. To solve this problem, we build performance models for computation and communication tasks in MoE and develop an optimal solution to determine the pipeline degree such that the iteration time is minimal. We conduct extensive experiments with 174 typical MoE layers and two real-world NLP models on a 64-GPU cluster. Experimental results show that our PipeMoE almost always chooses the best pipeline degree and outperforms state-of-the-art MoE training systems by 5%-77% in training time.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Prediction of Liver Disease using Machine Learning Models with PCA

Prediction of Liver Disease using Machine Learning Models wi...

引用

2021 International Conference on data analytics for Business and Industry, ICDABI 2021

作者： Al Telaq, Burair Hassan Hewahi, Nabil University of Bahrain Big Data Science and Analytics College of Science Sakir Bahrain University of Bahrain College of Information Technology Department of Computer Science Sakir Bahrain

ISBN: (纸本)9781665416566

The liver is one of the most vital organs of the human body. There are various disorders of liver that need early treatment by doctors. Early diagnosis and treating the patients are significant to reduce the risk. Healthcare system can benefit from various Machine Learning (ML) models to predict diseases in early stage. The aim of this study is to predict liver disease using different ML models applied on Indian Liver Patient dataset (ILPD). The models used on this work are Support Vector Machine (SVM), K-Nearest Neighbor (KNN), Random Forest (RF), Artificial Neural Network (ANN) and various versions of Ensemble Learning (EL). The results show that ensemble learning of KNN, RF, and SVM achieved the highest accuracy of 88%. Moreover, ensemble learning of KNN with RF obtained the highest True Positive Rate (TPR) of 99% on negative cases. In addition, this paper found that the Principal Component Analysis (PCA) technique has good impact to increase the accuracy. © 2021 IEEE.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Pretrained Sentence Embedding and Semantic Sentence Similarity Language Model for Text Classification in NLP

Pretrained Sentence Embedding and Semantic Sentence Similari...

引用

International Symposium on Artificial Intelligence and Signal Processing (AISP)

作者： V.Valli Mayil T.Ratha Jeyalakshmi Department of Computer Science and Engineering Koneru Lakshmaiah Education Foundation Vaddeswaram Andhra Pradesh India Department of Data Science & Analytics Jain University School of Sciences Banglore Karnataka

A key task in the field of Natural Language Processing (NLP) is determining semantic similarity between text sentences. Sentence pair modeling, Textual similarity and language modeling are few important tasks in NLP. Traditional machine learning algorithms require an enormous quantity of training data, but it is a timeconsuming process. Pre-trained models can be modified for a variety of downstream applications since they use methods for generically learning the characteristics of neural network topologies and language representations. Bidirectional Encoder Representations from Transformers- BERT & GPT are the popular architectures in NLP which enable to use minimal fine-tuning effort to produce effective results. In this work a fine-tuned BERT model that is suitable for semantic sentence similarity which predicts the entailment, neutral and contradictory categories of sentence pairs is presented. The fine-tuning feature promotes the training phase of model whichis widely effective across different types of semantic similarity models. The performance analysis of our system shows that the fine-tuned model reduces the number of neurons in the neural network there by reducing storage and time spent in expensive training task to create deep learning model.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Exploring Machine Learning Classifiers for Medical datasets

Exploring Machine Learning Classifiers for Medical Datasets

引用

2021 International Conference on data analytics for Business and Industry, ICDABI 2021

作者： Taha, Ranyah Alshakrani, Sara Hewahi, Nabil University of Bahrain Big Data Science and Analytics College of Science Sakhir Bahrain University of Bahrain College of Information Technology Department of Computer Science Sakhir Bahrain

ISBN: (纸本)9781665416566

This research paper studies different classification algorithms with the aim of finding the best Machine Learning (ML) classifiers for medical datasets to help Clinical Decision Support Systems (CDSS). This will help physicians in making more accurate and early diagnoses. Many studies used single medical dataset to evaluate ML classifiers but its unsatisfying and doesn't reflect the capability of the classifier on different medical datasets. In this research paper, eight ML classifiers experimented on five medical datasets gathered from UCI machine learning repository. The Random Forest (RF) classifier obtained the highest accuracy average with 83.822 %, also Logistic regression (LR), Decision Tree (DT) and Gradient Boosting showed similar performance with 81.666%, 81.364%, 81.034%, while AdaBoost and Gaussian Naïve Bayes (GNB) obtained 79.504% and 78.126%, Moreover, the last two classifiers in this research obtained the lower accuracy averages are K Nearest Neighbor (KNN) with 68.208% and Artificial Neural Network (ANN) with 55.49%. © 2021 IEEE.

关键词： Decision trees

来源：评论

学校读者我要写书评

暂无评论

Urban Sound Classification Using DNN, CNN LSTM a Comparative Approach

Urban Sound Classification Using DNN, CNN LSTM a Comparative...

引用

2021 International Conference on Innovation and Intelligence for Informatics, Computing, and Technologies, 3ICT 2021

作者： Bubashait, Mohamed Hewahi, Nabil Big Data Science and Analytics Master Program College of Science University of Bahrain Sakhier Bahrain University of Bahrain Department of Computer Science College of IT Sakhier Bahrain

ISBN: (纸本)9781665440325

Like air pollution, sound pollution has grown to be a major concern for city residents, designers, and developers. Detecting and recognizing sound types and sources in cities and suburban areas or any environment have become a necessity for the quality of life as well as security. In recent years, researchers have explored many models using Convolutional Neural Network (CNN), Long Short Term Memory (LSTM) Neural Network, and different combinations of these techniques, which produced promising results when combined with spectrogram images, or its different variations, to classify urban sounds. This research uses the DNN model performance as a baseline to compare the CNN and LSTM models' performance for classifying urban sound using Mel scale cepstral analysis (MEL) spectrum images using an open-source library called Librosa for sound processing. Models' performance was evaluated using the UrbanSound8k dataset. The CNN model underperformed with an accuracy rate of 87.15% and f1 score of 85.63% compared to both the DNN base model and the LSTM model. In contrast, comparing the LSTM model with CNN, LSTM shows better accuracy performance on test data with 90.15%, and f1 score of 90.15%. © 2021 IEEE.

关键词： Long short-term memory

来源：评论

学校读者我要写书评

暂无评论

Sentiment Analysis of Tweets during COVID-19 Pandemic Using BLSTM

Sentiment Analysis of Tweets during COVID-19 Pandemic Using ...

引用

2021 International Conference on data analytics for Business and Industry, ICDABI 2021

作者： Tareq, Alaa Hewahi, Nabil College of Science University of Bahrain Big Data Science and Analytics Sakhir Bahrain College of Information Technology University of Bahrain Department of Computer Science Sakhir Bahrain

ISBN: (纸本)9781665416566

social media is the new way for people to express and share their thoughts and it plays a huge role in spreading the anxiety widely during the pandemics and Twitter is one of the social media channels with high number of users and daily tweets. Clarifying and understanding what people thinks and their shared opinions during such hard times can reduce the burden on the health systems and entities and redirects the concerned entities by highlighting the areas to spread the awareness in. The aim of this study is to analyze and assess the sentiment of the Tweets shared during the COVID-19 pandemic and testing the Bidirectional Long Term Short Memory (BLTSM) of Recurrent Neural Network (RNN) in predicting the sentiment class which are Negative, Positive and Neutral. The results show slight difference between the positive and negative tweets which needs an attention to spread the awareness and hence, reduce the negativity. Furthermore, the BLSTM predicted the sentiment classes (Negative, Positive and Neutral) and obtained 86.15% accuracy rate. The high accuracy concludes that BLSTM can adapt and predict the sentiment of a text. © 2021 IEEE.

关键词： COVID-19

来源：评论

学校读者我要写书评

暂无评论

Parameter Efficient Fine Tuning Llama 3.1 for Answering Arabic Legal Questions: A Case Study on Jordanian Laws

Parameter Efficient Fine Tuning Llama 3.1 for Answering Arab...

引用

Computational Intelligence Approaches and Applications (ICCIAA), International Conference on

作者： Mohammed Fasha Bassam Hammo Bilal Sowan Husam Barham Esam Al-Nsour Business Intelligence and Data Analytics Department University of Petra Amman Jordan King Abdullah II School of Information Technology University of Jordan Amman Jordan Computer Science Department University of Zarqa Zarqa Jordan

ISBN: (数字)9798331523657

ISBN: (纸本)9798331523664

This study uses Jordanian law as a case study to explore the fine-tuning of the Llama-3.1 large language model for Arabic question-answering. Two versions of the model- Llama-3.1-8B-bnb-4bit and Llama-3.1-8B-Instruct-bnb-4bit-were fine-tuned using parameter-efficient fine-tuning (PEFT) with LoRA adapters and 4-bit quantized models, leveraging the Unsloth framework for accelerated and resource-efficient training. A custom dataset of 6000 legal question-answer pairs was curated from Jordanian laws and formatted into structured prompts. Performance was evaluated using the BLEU and the ROUGE metrics to compare the fine-tuned models to their respective base versions. Results demonstrated improved legal reasoning and accuracy while achieving resource efficiency through quantization and optimized fine-tuning strategies. This work underscores the potential of adapting large language models for Arabic legal domains and highlights effective techniques for fine-tuning domain-specific tasks.

关键词： Training Adaptation models Technological innovation Accuracy Text analysis Law Terminology Large language models Question answering (information retrieval) Tuning

来源：评论

学校读者我要写书评

暂无评论

BitmapAligner:Bit-Parallelism String Matching with MapReduce and Hadoop

引用

computers, Materials & Continua 2021年第9期68卷 3931-3946页

作者： Mary Aksa Junaid Rashid Muhammad Wasif Nisar Toqeer Mahmood Hyuk-Yoon Kwon Amir Hussain Department of Computer Science COMSATS University IslamabadWah47040Pakistan Department of Computer Science AIR University IslamabadAerospace and Aviation Campus Kamra43570Pakistan Department of Computer Science National Textile UniversityFaisalabad37610Pakistan Department of Industrial Engineering and Research Center for Electrical and Information Technology Seoul National University of Science and TechnologyNowon-GuSeoul01811Korea Data Science and Cyber Analytics Research Group Edinburgh Napier UniversityEdinburghEH114DYUK

Advancements in next-generation sequencer(NGS)platforms have improved NGS sequence data production and reduced the cost involved,which has resulted in the production of a large amount of genome *** downstream analysis of multiple associated sequences has become a bottleneck for the growing genomic data due to storage and space utilization issues in the domain of *** traditional string-matching algorithms are efficient for small sized data sequences and cannot process large amounts of data for downstream *** study proposes a novel bit-parallelism algorithm called BitmapAligner to overcome the issues faced due to a large number of sequences and to improve the speed and quality of multiple sequence alignment(MSA).The input files(sequences)tested over BitmapAligner can be easily managed and organized using the Hadoop distributed file *** proposed aligner converts the test file(the whole genome sequence)into binaries of an equal length of the sequence,line by line,before the sequence alignment *** Hadoop distributed file system splits the larger files into blocks,based on a defined block size,which is 128 MB by *** can accurately process the sequence alignment using the bitmask approach on large-scale sequences after sorting the *** experimental results indicate that BitmapAligner operates in real time,with a large number of ***,BitmapAligner achieves the exact start and end positions of the pattern sequence to test the MSA application in the whole genome query *** MSA’s accuracy is verified by the bitmask indexing property of the bit-parallelism extended shifts(BXS)*** dynamic and exact approach of the BXS algorithm is implemented through the MapReduce function of Apache ***,the traditional seeds-and-extend approach faces the risk of errors while identifying the pattern sequences’***,the proposed model resolves the largescale data challen

关键词： Next-generation sequencing multiple sequence alignment MapReduce Hadoop whole-genome big data bit-parallelism

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：