1,721,033 research outputs found
Venture into the unexplored: How developers view the use of agile tools to improve onboarding?
Get PDFEvery year thousands of software developers switch companies and teams and go through the onboarding process. It is an important activity that impacts long-term productivity and retention. Hence, software companies are continuously exploring new ways to improve the onboarding process. Onboarding involves tasks like communicating with mentors and peers to get an idea about the team's project. Agile tools which are primarily used for issue tracking can be useful to provide this information to onboarding members. In this paper, we collaborate with software developers to identify their views regarding the use of agile tools to quicken onboarding. We recursively designed an instrument of 26 questions and surveyed 9 developers who mentored newcomers during onboarding. The survey findings state that developers are interested to explore agile tools to improve the onboarding process. They also identified some challenges in this process and provided recommendations to mitigate those
Going Beyond Counting First Authors in Author Co-citation Analysis
Get PDFThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
Enhancing Stock Market Using Data Mining Analysis
Get PDFPredicting stock market trends is a complex challenge, owing to the volatility and intricacy of the market. Traditional methods often struggle to provide reliable forecasts in such dynamic environments. This study proposes an innovative approach that leverages machine learning, deep learning, sentiment analysis, and ensemble techniques to enhance stock market prediction. We comprehensively review existing literature in these domains, highlighting their strengths and weaknesses. Our methodology involves sourcing historical stock data from Yahoo Finance, stock-related comments, and news articles from Reddit and other News APIs. The data is preprocessed to train various models, including Long Short-Term Memory (LSTM) networks and Seasonal Autoregressive Integrated Moving Average with Exogenous Regressors (SARIMAX). Sentiment analysis is performed on the Reddit data and news articles using the FinBERT sentiment analyzer and incorporated into the models. Ensemble techniques like stacking, blending, and weighted averaging combine individual models' predictions. The proposed approach is evaluated using key metrics such as root mean squared error (RMSE), mean absolute error (MAE), and Sharpe ratio. Experimental results demonstrate the effectiveness and superiority of our approach over benchmark models such as random forest, support vector machines, and gradient boosting machines. The findings underscore the value of integrating machine learning, deep learning, sentiment analysis, and ensemble methods for stock market prediction. This research advances the field and provides valuable insights for future research and practical applications in financial decision-making
Comparative study of Energy using Apache Spark and Hadoop
Get PDFComputing power generation and its need has been increasing day by day due to the high utilization of processing power. This becomes a major challenge nowadays. Data centers account for around 2% of all global carbon emissions. In the US alone, 0.5% of total greenhouse gas emissions are attributed to data centers. That is about the same as the airline industry. Data centers and cloud servers are using more energy. This research study focuses on accommodating the demo workloads which depict the real-time scenario and make the stand-alone system as a single-user server. This represents the impact of single-user daily task and load on the servers. This study includes the potential future work to improve the data with respect to computing power. The machine learning model has been the major source of consuming computing power which also comes with the predefine libraries and its own impact on the System resources. The study has been conducted with all the main scenarios related to data centers as Linux has been used to be as close as the system performance and impact
Exploring Machine Learning and Statistical Learning Algorithms for Predicting Hypertension and Diabetes: A Comparative Study
Get PDFDiabetes and hypertension are two prevalent chronic diseases with significant health implications, including increased risk of cardiovascular events and mortality. Early detection and management of these conditions are crucial for preventing complications and improving patient outcomes. In this study, we leverage machine learning (ML) and deep learning (DL) techniques to predict diabetes and hypertension using the Behavioral Risk Factor Surveillance System (BFRSS) dataset. The dataset, despite its importance, exhibits class imbalance, particularly concerning the minority class representing diagnosed cases. To address this challenge, we employ Synthetic Minority Over-sampling Technique (SMOTE) to balance the data distribution and improve the performance of our models. ML algorithms, including Logistic Regression, Support Vector Machines (SVM), K-Nearest Neighbors (KNN), and ensemble methods like Random Forest and Gradient Boosting, were employed to analyze the dataset. Additionally, DL models, such as Convolutional Neural Networks (CNNs) and Multilayer Perceptrons (MLPs), were utilized to capture complex patterns in the data. Ensemble methods, including Random Forest and Gradient Boosting, along with deep learning algorithms, emerged as the top performers in our study, achieving remarkable results in predicting diabetes and hypertension. Specifically, these models demonstrated precision, recall, F1 score, and accuracy metrics of 0.77 on the test dataset. This underscores the effectiveness of ensemble learning techniques and deep learning architectures in accurately predicting chronic conditions such as diabetes and hypertension, showcasing their potential for enhancing medical diagnostic capabilities and improving patient care
Impact of performance on security: JWT Token
Get PDFMicroservices are gaining a lot of popularity among industry users. These services are collections of independent, small services to make a big application. Yet, the basic and strong fundamentals of microservices such as scalability, loose coupling, and automation, can further make them more vulnerable to security. Our research focuses on using the JWT Token to enhance the security of microservices for authentication purposes, considering performance. We have designed and implemented a simple security framework using JWT Token, interconnecting 3 small services using a composite microservice architectural pattern. We tried to emulate the real-world situation by sending 5000 sequential requests with security features disabled and with a JWT token, where different microservices are communicating with each other. Our findings show that the performance overhead of using JWT Token is 9%, which can be a valuable insight for practitioners. Index Terms—Microservices, REST APIs, SOA, distributed systems, cloud, authentication, REST, JWT
CloudScent: A Model for Code Smell Analysis in Open-Source Cloud
Get PDFThe low cost and rapid provisioning capabilities have made open-source cloud a desirable platform to launch industrial applications. However, as open-source cloud moves towards maturity, it still suffers from quality issues like code smells. Although, a great emphasis has been provided on the economic benefits of deploying open-source cloud, low importance has been provided to improve the quality of the source code of the cloud itself to ensure its maintainability in the industrial scenario. Code refactoring has been associated with improving the maintenance and understanding of software code by removing code smells. However, analyzing what smells are more prevalent in cloud environment and designing a tool to define and detect those smells require further attention. In this paper, we propose a model called CloudScent which is an open source mechanism to detect smells in open-source cloud. We test our experiments in a real-life cloud environment using OpenStack. Results show that CloudScent is capable of accurately detecting 8 code smells in cloud. This will permit cloud service providers with advanced knowledge about the smells prevalent in open-source cloud platform, thus allowing for timely code refactoring and improving code quality of the cloud platforms
Variations on the Author
Get PDF“Variations on the Author” discusses two of Eduardo Coutinho’s recent films (Um Dia na Vida, from 2010, and Últimas Conversas, posthumously released in 2015) and their contribution to the general question of documentary authorship. The director’s filmography is characterized by a consistent yet self-effacing form of authorial self-inscription: Coutinho often features as an interviewer that rather than express opinions propels discourses; an interviewer that is good at listening. This mode of self-inscription characterizes him as an author who is not expressive but who is nonetheless markedly present on the screen. In Um Dia na Vida, however, Coutinho is completely absent form the image, while Últimas Conversas, on the contrary, includes a confessional prologue that moves the director from the margins to the center of his films. This article examines the ways in which these works stand out in the filmography of a director who offers new insights into the notion of cinematic authorship
Appropriate Similarity Measures for Author Cocitation Analysis
Get PDFWe provide a number of new insights into the methodological discussion about author cocitation analysis. We first argue that the use of the Pearson correlation for measuring the similarity between authors’ cocitation profiles is not very satisfactory. We then discuss what kind of similarity measures may be used as an alternative to the Pearson correlation. We consider three similarity measures in particular. One is the well-known cosine. The other two similarity measures have not been used before in the bibliometric literature. Finally, we show by means of an example that our findings have a high practical relevance.information science;Pearson correlation;cosine;similarity measure;author cocitation analysis
- …
