Browsing by Author "Bashir, Maryam"

Now showing 1 - 2 of 2
  • No Thumbnail Available
    ItemMetadata only
    An intelligent linear time trajectory data compression framework for smart planning of sustainable metropolitan cities
    (Wiley, 2020-02-10) Bashir, Maryam; Ashraf, Jawad; Habib, Asad; Muzammil, Muhammad
    The urban road networks and vehicles generate exponential amount of spatio‐temporal big‐data, which invites researchers from diverse fields of interest. Global positioning system devices may transceive data every second thus producing huge amount of trajectory data. Subsequently, it requires optimized computing for various operations such as visualization and mining hidden patterns. This sporadically stored big‐data contains invaluable information, which is useful for a number of real‐time applications. Compression is a highly important, but knotty task. Optimized compression enables us achieve the desired results in efficient and effective manner by using minimum energy and computational resources without compromising on important information. We present two versions of a compression technique based on the points of intersections (PoI) of urban roads networks. Based on intelligent mining paradigm, we created a compressed lookup lexicon to store the PoIs of dynamically selected region of interests (ROI). An important feature of our lexicon is the key pattern, which is intelligently computed based on the relative geographic position of a spatial geodetic vertex with respect to Euclidean space origin in a given ROI. This compresses trajectories in linear time, making it feasible for mission critical real world applications. Our experimental dataset contained 959 547, 517 436, and 231 740 trajectories for Bikes, Cars, and Taxis, respectively. The Compr10 reduced these trajectories to 17 428, 11 084, and 6565, respectively. Results of Compr15 and Compr20 show promising results. We define the quality of the compression in context of the considered problem. The results show that the proposed technique achieved satisfactory quality of the compression.
  • No Thumbnail Available
    ItemMetadata only
    Predicting at-Risk Students at Different Percentages of Course Length for Early Intervention Using Machine Learning Models
    (IEEE, 2021-01-05) Adnan, Muhammad; Habib, Asad; Ashraf, Jawad; Mussadiq, Shafaq; Raza, Arsalan Ali; Abid, Muhammad; Bashir, Maryam; Khan, Sana Ullah
    Online learning platforms such as Massive Open Online Course (MOOC), Virtual Learning Environments (VLEs), and Learning Management Systems (LMS) facilitate thousands or even millions of students to learn according to their interests without spatial and temporal constraints. Besides many advantages, online learning platforms face several challenges such as students’ lack of interest, high dropouts, low engagement, students’ self-regulated behavior, and compelling students to take responsibility for settings their own goals. In this study, we propose a predictive model that analyzes the problems faced by at-risk students, subsequently, facilitating instructors for timely intervention to persuade students to increase their study engagements and improve their study performance. The predictive model is trained and tested using various machine learning (ML) and deep learning (DL) algorithms to characterize the learning behavior of students according to their study variables. The performance of various ML algorithms is compared by using accuracy, precision, support, and f-score. The ML algorithm that gives the best result in terms of accuracy, precision, recall, support, and f-score metric is ultimately selected for creating the predictive model at different percentages of course length. The predictive model can help instructors in identifying at-risk students early in the course for timely intervention thus avoiding student dropouts. Our results showed that students’ assessment scores, engagement intensity i.e. clickstream data, and time-dependent variables are important factors in online learning. The experimental results revealed that the predictive model trained using Random Forest (RF) gives the best results with averaged precision =0.60%, 0.79%, 0.84%, 0.88%, 0.90%, 0.92%, averaged recall =0.59%, 0.79%, 0.84%, 0.88%, 0.90%, 0.91%, averaged F-score =0.59%, 0.79%, 0.84%, 0.88%, 0.90%, 0.91%, and average accuracy =0.59%, 0.79%, 0.84%, 0.88%, 0.90%, 0.91% at 0%, 20%, 40%...