Using a Model for Learning and Memory to Simulate Learner Response in Spaced Practice

Web applications are crucial role for daily user activities such as online banking, online shopping and searching. It is important to ensure the reliability and web application testing has been used in finding various faults in order to improve the quality of reliable web services. Among test cases generation approaches, user session based testing is an approach to create test cases with real user data. However, real user data usage is extremely large and executing all the test cases can be time consuming in practice. This paper describes the test cases reduction approach for analyzing and replaying the large number of test cases generated from user session data. The entropy gain theory is applied in test cases reduction process to get the best test suite that covers all user accesses of web application. To evaluate the effectiveness of proposed method, the analytical results are described in terms of URLs coverage, reduction time and test cases reduction rate.

Large language models (LLMs) are increasingly used for mental health, yet safety evaluations rely primarily on small, simulation-based benchmarks removed from real-world language. We replicate four published safety evaluations assessing suicide risk handling, harmful content generation, and jailbreak resistance for general-purpose frontier models and a purpose-built mental health AI. We then conduct an ecological audit of 20,000 real user conversations with the purpose-built system, which includes layered safeguards for suicide and non-suicidal self-injury (NSSI). The purpose-built AI was significantly less likely than general-purpose LLMs to produce harmful content across suicide/NSSI (.4-11.27% vs 29.0-54.4%), eating disorder (8.4% vs 54.0%), and substance use (9.9% vs 45.0%) benchmarks. In real user data, clinician review found zero suicide-risk cases without crisis resources. Three NSSI mentions (.015%) lacked intervention, implying a .38% lower-bound false negative rate. Findings support the utility of ecological audits for safety estimation.

Aim: This study aims to improve science and social studies learning outcomes of third-grade students at SDN 2 Panarung through the implementation of a simple experiment-based Project-Based Learning (PjBL) model that accommodates kinesthetic and hyperactive learning characteristics. Method: The research was conducted using Classroom Action Research with the Kemmis and McTaggart model in two cycles involving 23 third-grade students. Data were collected through learning outcome tests, observations, and documentation, then analyzed descriptively using gain scores and narrative qualitative analysis. Result and Discussions: The findings indicate a significant improvement in student learning outcomes across cycles. The average score increased from 32.17 in the pre-cycle to 56.09 in Cycle I and 87.83 in Cycle II. The N-Gain analysis showed a moderate improvement in Cycle I (0.35) and a high improvement in Cycle II (0.84). Learning mastery also improved from 0% in the pre-cycle to 47.8% in Cycle I and reached 100% in Cycle II. The improvement in Cycle II was supported by the use of simpler and more concrete experiments that aligned with students’ developmental levels and learning styles. Conclusion, The implementation of simple experiment-based PjBL proved effective in enhancing learning outcomes, encouraging active participation, and increasing students’ understanding and motivation. This learning model is recommended for use in classrooms with similar student characteristics.

The development of the global tourism industry rapidly makes the importance of English communication skill at tourism workplace. State Polytechnic of Lampung has provided English subjects 1, 2, and 3 for students. Unfortunately, the implementation has not given any significant contribution to improve students’ English communication skills especially English for tourism. English instructional materials did not fulfill students’ needs and stakeholders’ needs and standard English competences in tourism workplace. The limitation of learning source become a learning problem for students due to limitation of time to practice English out site of classroom without teacher’s guidance. Therefore, it is important to develop a kind of learning model which independence resource that gives a solution to solve the learning problems. Developing model of English learning for tourism based on hypermedia was intended to give an alternative independence learning resource to improve competences in English for Tourism based on hypermedia. The developing of English learning model base on hypermedia applied research and development (R&D) model used a combination of Dick and Carrey model, Hannafin peck Model and Allen model. There were three stages in developing model of English for tourism based on hypermedia. The first step was conducting needs analysis; identifying instructional goals, conducting instructional analysis, and analyzing learners and contexts. The second step was designing material for learning based on hypermedia that was electronic module; writing instructional objectives, developing assessment instruments, and developing learning strategies. The third steps were developing learning material and implementation; developing and selecting instructional materials, implementing and conducting formative evaluation and revision. The results of the formative evaluation from subject matter experts (instructional, media and English for tourism), one to one evaluation, showed that model of English learning for tourism based on hypermedia generally was categorized very good (average score: 89,31). One to one evaluation by three learners was also and small group evaluation by eight learners which evaluated on clarity, impact on learners and feasibility showed that design model of English learning for tourism based on hypermedia was categorized very good. Field trial that was conducted to 37 students indicated that model design of English learning based on hypermedia for tourism was good and suitable with learning objective, easy to use and understand, demonstrate learning experiences, sufficient for English materials, useful, assessment measure learning outcome, feedback, interesting and engaging learning, motivating, and collaboration between students. The t-test paired sample test showed that design of English for tourism was effective to be used as learning resources with average pretest score was 58,05 and average post test score was 70,92 with gain score was 12,86, while the significant (2-tailed) was 0,000 < 0,05.

This study aims to develop a contextual learning model for the topic of projectile motion by integrating Hombo Batu activities as a form of local wisdom from the South Nias community. Hombo Batu, or the traditional stone-jumping ritual, is a physical activity rich in two-dimensional kinematics elements, such as initial velocity, angle of elevation, and parabolic trajectory. The research employed the Research and Development (R&D) method based on Sugiyono's development model, which includes the following stages: identifying potential and problems, data collection, product design, validation, revision, and limited trials. Validation results from subject matter experts and media experts indicated that the developed learning model achieved a high level of validity (Aiken’s V > 0.85), categorized as “highly feasible.” A limited trial conducted with junior high school students in South Nias showed that the model effectively improved students' understanding of projectile motion, with an average N-Gain score of 0.56 (moderate to high category). Additionally, the implementation of the model led to increased learning motivation and greater appreciation of local culture among the students. This model not only enhances conceptual understanding in physics but also strengthens students’ character and cultural identity through a learning approach that is contextual, meaningful, and applicable. Therefore, the Hombo Batu-based contextual learning model is recommended as an innovative alternative for teaching physics, particularly in regions rich in local cultural values.

Carpooling is a flexible shared transportation system which can effectively reduce the vehicle numbers and fuel consumption. Although many carpooling systems have been proposed, most of them lack practicality, veracity, and efficiency. In this paper, we propose a new useful variant model of the long-term carpooling problem which involves multiple origins and one destination. Such problems commonly occur in a wide number of carpooling situations in real-world scenarios. Our work is motivated by the practical needs to solve environmental pollution, parking problems, traffic jams and low utilization of resources. A Tabu search algorithm is proposed in this paper to solve the carpooling problem. The proposed algorithm aims at a wide range of passenger distribution and routing problems. The computational results based on real world user data show the effectiveness of the proposed algorithm. Moreover, we developed a mobile application based on our carpooling model.

Harassing and fraud calls have spread like viruses in people's lives, many researchers have proposed some solutions to abnormal phone detection. However, most of these methods are passive detection, cannot give accurate prediction in time. In this work, we worked with operators to obtain a volume of real telecom user data and extract a series of comprehensive features. We propose an integration method of classifiers for abnormal phone detection by applying the machine learning algorithm on the data with unbalance and ‘dirty data’. Especially, we use bootstrap sampling method and voting strategy to reduce the false prediction of classier due to noise data. The experimental result shows the effectiveness of our method in contrast with traditional classification algorithm.

ns2 Simulator has been widely used to simulate various scenarios related to wired and wireless medium. However, transfer of real time user data has not been implemented as researchers are more concerned with the output of the system in terms of range of values of particular parameter desired or else. This framework besides running the default simulation as done by ns2 adds the capability to allow transfer of user data and allows user to save at the other end or destination the data received. The user data may be multimedia data or any data (plain text), document file, binary file or raw data. Simulations result show that actual user data received may be same or different, due packet error or noise, respectively depending upon the network topologies used. The user data transfer has been implemented in UDP (CBR traffic).

This paper proposes mobility prediction technique via Markov Chains with an input of user’s mobile data traces to predict the user’s movement in wireless network. The main advantage of this method is prediction will give knowledge of user’s movement in advance even in fast moving vehicle. Furthermore, the information from prediction result will be use to assist handover procedure by reserve resource allocation in advance in vehicular network. This algorithm is simple and can be computed within short time; thus the implementation of this technique will give the significant impact especially on higher speed vehicle. Finally, an experiment is performed using real mobile user data traces as input for Markov chain to predict next user movement. To evaluate the effectiveness of the proposed method, MATLAB simulations are carried out with several users under same location zone. The results show that the proposed method predicts have good performance which is 30% of mobile users achieved 100% of prediction accuracy.

Tracking dietary intake is an important task for health management especially for chronic diseases such as obesity, diabetes, and cardiovascular diseases. Given the popularity of personal hand-held devices, mobile applications provide a promising low-cost solution to tackle the key risk factor by diet monitoring. In this work, we propose a photo based dietary tracking system that employs deep-based image recognition algorithms to recognize food and analyze nutrition. The system is beneficial for patients to manage their dietary and nutrition intake, and for the medical institutions to intervene and treat the chronic diseases. To the best of our knowledge, there are no popular applications in the market that provide a high-performance food photo recognition like ours, which is more convenient and intuitive to enter food than textual typing. We conducted experiments on evaluating the recognition accuracy on laboratory data and real user data on Singapore local food, which shed light on uplifting lab trained image recognition models in real applications. In addition, we have conducted user study to verify that our proposed method has the potential to foster higher user engagement rate as compared to existing apps based dietary tracking approaches.

Voice assistants are deployed widely and provide useful functionality. However, recent work has shown that commercial systems like Amazon Alexa and Google Home are vulnerable to voice-based confusion attacks that exploit design issues. We propose a systems-oriented defense against this class of attacks and demonstrate its functionality for Amazon Alexa. We ensure that only the skills a user intends execute in response to voice commands. Our key insight is that we can interpret a user's intentions by analyzing their activity on counterpart systems of the web and smartphones. For example, the Lyft ride-sharing Alexa skill has an Android app and a website. Our work shows how information from counterpart apps can help reduce dis-ambiguities in the skill invocation process. We build SkilIFence, a browser extension that existing voice assistant users can install to ensure that only legitimate skills run in response to their commands. Using real user data from MTurk (N = 116) and experimental trials involving synthetic and organic speech, we show that SkillFence provides a balance between usability and security by securing 90.83% of skills that a user will need with a False acceptance rate of 19.83%.

The TaskTracer system allows knowledge workers to define a set of activities that characterize their desktop work. It then associates with each user-defined activity the set of resources that the user accesses when performing that activity. In order to correctly associate resources with activities and provide useful activity-related services to the user, the system needs to know the current activity of the user at all times. It is often convenient for the user to explicitly declare which activity he/she is working on. But frequently the user forgets to do this. TaskTracer applies machine learning methods to detect undeclared activity switches and predict the correct activity of the user. This paper presents TaskPredictor2, a complete redesign of the activity predictor in TaskTracer and its notification user interface. TaskPredictor2 applies a novel online learning algorithm that is able to incorporate a richer set of features than our previous predictors. We prove an error bound for the algorithm and present experimental results that show improved accuracy and a 180-fold speedup on real user data. The user interface supports negotiated interruption and makes it easy for the user to correct both the predicted time of the task switch and the predicted activity.

The discovery of a person's personally important places involves obtaining the physical locations for a person's places that matter to his daily life and routines. This problem is driven by the requirements from emerging location-aware applications, which allow a user to pose queries awl obtain, information in reference, to places, e.g., ''home", ''work" or ''Northwest Health Club". It is a challenge to map from physical locations to jxtrsonally meaningful places because GPS tracks are continuous data both spatially and temporally, while most existing data mining techniques expect discrete data. Previous work has explored algorithms to discover personal places from location data. However, they all have limitations. Our work proposes a two-step approach that discretized continuous GPS data into places and learns important places from the place features. Our approach was validated using real user data and shown to have good accuracy when applied in predicting not only important and frequent places, but also important and not so frequent places.

#TurkihTweets is a benchmark dataset for the task of correcting the user misspellings, with the purpose of introducing the first public Turkish dataset in this area. \#TurkihTweets provides correct/incorrect word annotations with a detailed misspelling category formulation based on the real user data. We evaluated four state-of-the-art approaches on our dataset to present a preliminary analysis for the sake of reproducibility.

Digital transformation in education requires media that sustain student engagement in blended settings. This research seeks to design and develop a Moodle-based Learning Management System for Informatics at Vocational High School 1 Koto Baru and to verify its quality and usability for routine classroom use. The study uses a phased development approach (Analyse, Design, Develop, Implement, Evaluate). Participants were 44 tenth-grade learners and 2 instructors; quality was reviewed by 2 subject-matter experts and 3 media experts. Data came from structured questionnaires and were analysed descriptively in %. Findings show content quality is highly valid at 90% overall (89% content indicators, 92% instructional). Media quality is likewise highly valid at 91% (90% ease of use, 93% visual appeal). Classroom applicability is strong, with practicality of 92% for instructors and 92% for learners. Gamification mechanisms (points, levelling up, badges, and leaderboards) provide quick feedback, structured progress tracking, and support for learning orchestration. Overall, the product meets highly valid and practical criteria and is ready for adoption as a modular, trackable Moodle-based course for Informatics at the vocational high school level. Future work includes quasi-experimental tests of learning outcomes, replication across schools and subjects, performance evaluation under high access loads, and development of adaptive learning analytics.