Genetic Algorithm-Based Support Vector Machine Approach for Intelligent Usability Assessment of Mobile Learning Applications

Core Concepts

A novel approach using genetic algorithm-based support vector machine to extract and score key usability features of a mobile learning application as per user requirements.

Abstract

The study proposes a novel approach for intelligent usability assessment of mobile learning applications. It involves the following key steps: Collecting user feedback and requirements regarding the usability of a mobile learning application (in this case, the LMS-IUB application) through a questionnaire-based survey. Extracting key usability features (efficiency, effectiveness, ease of use, learnability, memorability, cognition, and consistency) from the user feedback data. Scoring the extracted usability features using a genetic algorithm-based approach. The genetic algorithm is used to optimize the scoring of the features based on user requirements. Applying a GA-based support vector machine model to cluster and classify the usability features based on their importance and quality. Ranking the usability features into categories like "very good", "good", "fair", and "average" based on the scoring and classification results. The proposed approach provides a systematic and intelligent way to assess the usability of mobile learning applications from the user's perspective. It helps identify the key usability issues and prioritize them for improving the overall user experience.

Stats

The study used a sample of 200 students from the Islamia University of Bahawalpur to collect user feedback on the usability of the LMS-IUB mobile application.

Quotes

"In the field of human computer interaction (HCI), the usability assessment of m-learning (mobile-learning) applications is a real challenge." "There is neither any theory nor any tool available to measure or assess a user's perception and assessment of usability features of a m-learning application for the sake of ranking of the graphical user interface of a mobile application in terms of a user's acceptance and satisfaction."

Key Insights Distilled From

A Genetic Algorithm-Based Support Vector Machine Approach for Intelligent Usability Assessment of m-Learning Applications

by Muhammad Asg... at arxiv.org 04-26-2024

https://arxiv.org/pdf/2404.16043.pdf

Deeper Inquiries

How can the proposed approach be extended to assess the usability of mobile applications in other domains beyond education?

The proposed approach of using a Genetic Algorithm-Based Support Vector Machine for intelligent usability assessment of m-learning applications can be extended to assess the usability of mobile applications in various domains beyond education by adapting the methodology to suit the specific requirements and characteristics of those domains. Here are some ways in which the approach can be extended: Customization of Features: Different domains may have unique usability features that are critical for assessing the quality of mobile applications. The approach can be extended by customizing the feature selection process to include domain-specific usability attributes. Diverse Datasets: Collecting diverse datasets from users in different domains will be essential to train the machine learning models effectively. This will involve understanding the specific needs and preferences of users in those domains. Expert Input: In domains where domain-specific knowledge is crucial, involving experts from those fields to provide input on the usability features and their importance can enhance the accuracy of the assessment. Evaluation Metrics: The evaluation metrics used to assess the usability of mobile applications may need to be tailored to each domain to ensure that the assessment is relevant and meaningful. Testing and Validation: Extensive testing and validation of the approach across different domains will be necessary to ensure its effectiveness and reliability in assessing usability in diverse contexts. By incorporating these strategies and adapting the approach to the specific requirements of different domains, the proposed methodology can be successfully extended to assess the usability of mobile applications beyond education.

What are the potential limitations of using a genetic algorithm-based approach for usability feature scoring, and how can they be addressed?

While using a genetic algorithm-based approach for usability feature scoring offers several advantages, there are also potential limitations that need to be considered. Some of these limitations include: Computational Complexity: Genetic algorithms can be computationally intensive, especially when dealing with large datasets or complex feature selection processes. This can lead to longer processing times and resource constraints. Optimization Challenges: Genetic algorithms may struggle to find the optimal solution in highly nonlinear or non-convex optimization problems. This can result in suboptimal feature selection and scoring. Overfitting: There is a risk of overfitting the model to the training data, especially if the genetic algorithm is not properly tuned or if the dataset is not representative of the overall population. Interpretability: Genetic algorithms can produce complex models that are difficult to interpret, making it challenging to understand the reasoning behind the feature scoring decisions. These limitations can be addressed by: Optimizing Parameters: Fine-tuning the parameters of the genetic algorithm and the machine learning models can help improve performance and reduce computational complexity. Feature Engineering: Conducting thorough feature engineering and preprocessing of the data can help improve the quality of input features and enhance the performance of the genetic algorithm. Cross-Validation: Implementing cross-validation techniques can help prevent overfitting and ensure the generalizability of the model to unseen data. Model Interpretation: Using techniques such as feature importance analysis and model visualization can aid in interpreting the results and understanding the impact of each feature on usability scoring. By addressing these limitations, the genetic algorithm-based approach can be optimized for usability feature scoring and deliver more accurate and reliable results.

How can the user feedback collection process be further improved to capture more nuanced and comprehensive insights on the usability of mobile learning applications?

Improving the user feedback collection process is crucial for capturing nuanced and comprehensive insights on the usability of mobile learning applications. Here are some strategies to enhance the user feedback collection process: Diverse Feedback Channels: Offer multiple channels for users to provide feedback, such as surveys, interviews, focus groups, and in-app feedback forms. This allows users to choose the method that best suits their preferences. Real-Time Feedback: Implement real-time feedback mechanisms within the mobile application to capture immediate user reactions and suggestions while using the app. In-Depth Interviews: Conduct in-depth interviews with a diverse group of users to gain deeper insights into their experiences, challenges, and suggestions for improving usability. Usability Testing: Conduct usability testing sessions with real users to observe their interactions with the mobile application and gather feedback on specific usability features. Feedback Analysis Tools: Utilize feedback analysis tools and sentiment analysis techniques to extract valuable insights from user comments and reviews. Continuous Improvement: Establish a feedback loop to continuously collect, analyze, and implement user feedback to enhance the usability of the mobile learning application. Engagement Incentives: Offer incentives or rewards for users who provide detailed and constructive feedback to encourage active participation in the feedback collection process. Accessibility: Ensure that the feedback collection process is accessible and user-friendly to encourage maximum participation from users with diverse backgrounds and abilities. By implementing these strategies, the user feedback collection process can be further improved to capture nuanced and comprehensive insights on the usability of mobile learning applications, leading to enhanced user experiences and overall application quality.