Hello folks,           Let's test the following functions:       def double(arg):     """ This functi... Tagged with python, algorithms, database, machinelearning.

Calling by value & by reference in Python

Hello Geeks  In a simple way, distributed systems is a Shared Architecture to achieve three aspects,... Tagged with distributedsystems, database, bigdata, cloud.

Distributed Systems in Terms of Scalability, Fault tolerance and Latency

Hi there!  Let's jump into the steps directly to save your time 🙂  ON Upwork, clients need to work... Tagged with career, tutorial, productivity, beginners.

Get your proposal winning: Bits of Advice from TOP RATED FREELANCER on Upwork

Big Data Stream processing engines such as Apache Flink use windowing techniques to handle unbounded streams of events. Gathering all pertinent input within a window is crucial for event-time windowing since it affects how accurate results are. A significant part of this process is played by watermarks, which are unique timestamps that show the passage of events in time. However, the current watermark generation method in Apache Flink, which works at the level of the input stream, tends to favor faster sub-streams, resulting in dropped events from slower sub-streams. In our analysis, we found that Apache Flink’s vanilla watermark generation approach caused around 33% loss of data if 50% of the keys around the median are delayed. Furthermore, the loss surpassed 37% when 50% of random keys are delayed.In this paper, we present a novel strategy called keyed watermarks to overcome data loss and increase the accuracy of data processing to at least 99% in most cases. We enable separate progress tracking by creating a unique watermark for each logical sub-stream (key). In our study, we outline the architectural and API changes necessary to implement keyed watermarks and discuss our experience in extending Apache Flink’s enormous code base. Additionally, we compare the effectiveness of our strategy against the conventional watermark generation method in terms of the accuracy of event-time tracking.

Keyed Watermarks: A Fine-grained Tracking of Event-time in Apache Flink

The exploration of sentiment analysis in multilingual contexts, particularly through the integration of deep learning techniques and knowledge graphs, represents a significant advance in language processing research. This study specifically concentrates on the Arabic language, addressing the challenges presented by its morphological complexity. While the primary focus is Arabic, the research also includes a comprehensive review of related work in other languages such as Bangla and Chinese. This contextualizes the challenges and solutions found in Arabic sentiment analysis within a broader multilingual landscape. Utilizing pre-trained language models like BERT, the research has achieved noteworthy improvements in sentiment analysis accuracy and efficiency, particularly for the Arabic language. The integration of knowledge graphs stands out as a crucial innovation, offering essential contextual insights and mitigating the limitations posed by sparse labeled datasets in Arabic, a language less resourced compared to English. The findings of this study highlight the effectiveness of tailored BERT models for Arabic sentiment analysis, revealing the vast potential and inherent challenges of employing knowledge graphs and large language models for a deeper, more nuanced understanding. The future direction of this research includes enhancing these methods with cutting-edge machine learning techniques, aiming to further refine sentiment analysis processes and knowledge graph construction with a focus on Arabic within a multilingual framework.

Sentiment Analysis for Arabic Product Reviews Using LLMs and Knowledge Graphs

This study introduces a novel approach for detecting malware using cutting-edge machine Learning with PySpark, which targets the increasingly complex cyber threats that conventional security measures struggle to counter. Our research aims to create a highly accurate system for detecting harmful network activities by examining network traffic patterns, including the duration and frequency of connections, the amount of data transferred, and the involved endpoints. We utilized a comprehensive dataset from Kaggle and performed thorough data preprocessing and feature selection to optimize our models' performance. The research employed a variety of algorithms such as Logistic Regression, Random Forest, Decision Tree, and Naive Bayes. Each model was fine-tuned and evaluated based on its recall in distinguishing between safe and dangerous software. Our work seeks to make a significant impact in the field of cybersecurity by presenting an effective machine learning-driven tool for accurate malware detection, thereby improving network security.

Advanced Network Malware Detection: Integrating PySpark and Machine Learning Techniques

This paper presents a comprehensive methodology for gender detection using hand palm images, leveraging image processing techniques and PySpark for scalable and efficient processing. The approach encompasses a meticulous image preprocessing pipeline, incorporating essential stages like grayscale conversion, the application of the Difference of Gaussians (DoG) filter, and adaptive histogram equalization. This approach not only refines image features but also ensures scalability, accommodating large datasets seamlessly. After preprocessing of hand images, the VGG19 model is employed as a powerful feature extractor, utilizing its pre-trained convolutional layers to capture high-level abstract features from hand palm images. The final step was to train and fit the tabular data and extracted features on the Multilayer Perceptron (MLP) model, which is a feedforward neural network for classification tasks in PySpark. The MLP model, with its ability to learn complex patterns, proves effective in accurately predicting gender categories based on the extracted features. Finally, the model results were 94% accuracy for training and 87% testing which indicates that the model is generalizing well to unseen data. Also, another evaluation metrics were used like Precision 0.87, Recall 0.87, F1 Score 0.87 and AUC 0.92.

Gender Detection from Hand Palm Images: A PySpark-Driven Approach with VGG19 Feature Extraction and MLP Classification

This paper introduces a groundbreaking enhancement to image captioning through a unique approach that harnesses the combined power of the Vision Encoder-Decoder model. By leveraging the Swin Transformer as the image encoder and RoBERTa as the text decoder, this architecture seamlessly integrates cutting-edge technologies in computer vision, natural language preprocessing, and big data. To demonstrate the efficacy of this innovative methodology, a rigorous evaluation has been conducted on the Flickr30k dataset allows for comprehensive examination and comparison of the models' performance against current state-of-the-art methods. Further-more, the explanation of the Flickr30k dataset expands the evaluation's reach and effectively addresses the ever-changing demands of image captioning challenges. By broadening the analysis to include Flickr30k, this study makes a valuable contribution to our understanding of the models' ability to adapt to diverse datasets, making it more suitable for real-world applications. The results of the evaluation on Flickr30k showcase the models' versatility and effectiveness in generating accurate and contextually relevant captions. By combining the power of Swin Transformer and RoBERTa, the proposed architecture not only outperforms established benchmarks but also solidifies it's position as a robust solution for tackling the complexities inherent in image captioning challenges.

Image Captioning Using Vision Encoder Decoder Model

Just a moment...

Ever since diving into the world of data, I’ve been captivated by its power to reveal… · Experience: Digital HUB (D-HUB) · Education: Nile University · Location: Cairo · 500+ connections on LinkedIn. View Tawfik Yasser’s profile on LinkedIn, a professional community of 1 billion members.

Tawfik Yasser - Digital HUB (D-HUB) | LinkedIn

Read writing from Tawfik Yasser on Medium. M.Sc. in Big Data and Streaming | Data Engineering. Every day, Tawfik Yasser and thousands of other voices read, write, and share important stories on Medium.

Tawfik Yasser – Medium

Ready to unlock the full potential of keyed watermarks? For those craving an in-depth dive, our complete research paper unveils everything—from our innovative component design to the striking results…

Revolutionizing Stream Processing: How Keyed Watermarks Enhance Apache Flink’s Accuracy

Low watermarks are a mechanism for tracking the progress of event time in a stream. In this blog post, I’ll take a closer look at watermark inner workings in Apache Flink, focusing on how they are…

Unraveling the Mystery of Watermarks in Apache Flink: An Internal Perspective

Stack Overflow | The World’s Largest Online Community for Developers

User Tawfik Yasser

Apache Kafka - Arabic

كيف تبدء الـ Freelancing؟

Hadoop The Definitive Guide | YARN  | Arabic

Hadoop The Definitive Guide | HDFS  | Arabic

Hadoop The Definitive Guide | MapReduce | Arabic

Hadoop The Definitive Guide | Meet Hadoop | Arabic

I publish research papers, share articles on Medium, educational videos on YouTube, share pictures on Instagram, and more! 🚀

I worked as a teaching assistant at Nile University