Predicting Media Impact: A Machine Learning Framework for Optimizing Corporate Communication Strategies in Architectural Practices

International Journal of Computational and Experimental Science and Engineering, Год журнала: 2025, Номер 11(1)

Опубликована: Фев. 8, 2025

The research investigates the role of media relations and corporate communications strategies architectural firms that conventionally pursue PR methodologies data-driven approaches have evolved. This has led to conduct studies use qualitative insights coupled with predictive modelling. These are used examine how companies evolving their approach in digital age. study ten leading architecture firms, assessing communication effectiveness through interviews, content analysis, social metrics. further predicts stakeholder engagement impact by applying machine learning models- Random Forest LSTM networks an accuracy 85%. Key findings include drivers based on sentiment, share ability, timing significant. demonstrated can drive strategic decision-making, optimize public relations, improve engagement. Moreover, provides easily scalable framework for forecasting purposes different markets. Further, it shows promise AI-driven strategies. Combining theory advanced analytics, this benefit from increasingly nature relations. been a major need proactive reputation management distribution. It enables others better adapt changing waves response maximal positive

Язык: Английский

---

Depression Sentiment Analysis using Machine Learning Techniques:A Review

International Journal of Computational and Experimental Science and Engineering, Год журнала: 2025, Номер 11(1)

Опубликована: Фев. 20, 2025

Depression is one of the habitual psychological well-being diseases and a significant number depressed individuals end their lives. People suffering from depression don’t ask for help doctors due to hesitation or unawareness about that causes delay in diagnosis treatment. A lot people share opinions emotions on social networking sites. Several studies site posts related rely upon Facebook, Twitter, Blogs, other networks because they recording behavioral attributes which are person’s thinking, socialization, communication, etc. Datasets various sites useful sentiment analysis. Various machine learning deep techniques like Naïve Bayes, maximum entropy, Support Vector Machine (SVM), Decision Tree classifiers neural networks, recurrent have been used detection. This paper presents review analysis performed media platforms detection The datasets utilized also discussed. comparative existing work area provided get clear understanding used. Finally, challenges future can be done field discussed

Язык: Английский

---

Practical Research on Project-Based Learning (PBL) in Film and Television Production in Xiamen Vocational Education

International Journal of Computational and Experimental Science and Engineering, Год журнала: 2025, Номер 11(1)

Опубликована: Фев. 23, 2025

The film and television industry plays a crucial role in the development of global cultural sector. In recent years, vocational education field has experienced rapid growth China. However, current talent training model for this profession fails to meet demands fast-paced lacks effective support its advancement. Project-based learning is student-centered teaching approach that employs authentic projects as primary medium learning. This study presents an empirical investigation conducted college Xiamen, where project-based was incorporated into production courses assess effectiveness. findings research demonstrate implementation context viable. comparison traditional didactic instruction, significantly enhances students' motivation learn, practical skills, critical thinking abilities, teamwork abilities. Consequently, it holds significant value cultivating applied talents.

Язык: Английский

---

A Graph-Based and Pattern Classification Approach for Kannada Handwritten Text Recognition Under Struck-Out Conditions

International Journal of Computational and Experimental Science and Engineering, Год журнала: 2025, Номер 11(1)

Опубликована: Март 4, 2025

This research focuses on the processing and identification of handwritten Kannada text, particularly under struck-out conditions. The database considered in this study comprises data. When such a is processed using optical character recognition (OCR)-based digital systems, output may often be an unrecognizable format. To address issue, model has been developed incorporating pattern classification graph-based method for text identification. For classification, feature extraction performed two different classes with support vector machines (SVMs) classifier. In approach, strokes are analyzed shortest path algorithm. handle zigzag or wavy all possible paths strike-out identified, suitable features extracted further processing. synthesized/recovered inpainting cleaning to ensure recovery. proposed methodology tested both trained untrained datasets script. Performance evaluation was conducted three parameters: precision, F1 score, accuracy.

Язык: Английский

---

Machine Learning Framework for Detecting Fake News and Combating Misinformation Spread on Facebook Platforms

International Journal of Computational and Experimental Science and Engineering, Год журнала: 2025, Номер 11(2)

Опубликована: Апрель 13, 2025

The spread of fake news on social media platforms like Facebook threatens societal harmony and undermines the reliability information. To address this issue, research employs machine learning techniques to construct a robust scalable framework for detecting news. Using well-curated dataset labeled posts containing both authentic news, study ensures balanced representation effective learning. Textual data was transformed into numerical features through Term Frequency-Inverse Document Frequency (TF-IDF) preprocessing, enabling seamless integration with algorithms. A variety classification models, including Support Vector Machines (SVM), Logistic Regression, Gradient Boosting, Random Forest, were trained evaluated. Six performance evaluations precision, accuracy, F1 score, recall, Matthews Correlation Coefficient (MCC), area under Receiver Operating Characteristic (ROC) curve—were used measure model effectiveness. results highlighted Boosting as most algorithm, achieving superior accuracy overall performance. This demonstrates capability automate detection misinformation, offering efficient solution preserving content credibility Facebook. contributes significantly broader effort combating ensuring dissemination reliable information, safeguarding public trust

Язык: Английский

---