Exploring jump back behavior patterns and reasons in e-book system

With the increasing use of digital learning materials in higher education, the accumulated operational log data provide a unique opportunity to analyzing student learning behaviors and their effects on student learning performance to understand how students learn with e-books. Among the students’ reading behaviors interacting with e-book systems, we find that jump-back is a frequent and informative behavior type. In this paper, we aim to understand the student’s intention for a jump-back using user learning log data on the e-book materials of a course in our university. We at first formally define the “jump-back” behaviors that can be detected from the click event stream of slide reading and then systematically study the behaviors from different perspectives on the e-book event stream data. Finally, by sampling 22 learning materials, we identify six reading activity patterns that can explain jump backs. Our analysis provides an approach to enriching the understanding of e-book learning behaviors and informs design implications for e-book systems.

Exploring for Route Preferences of Subway Passengers Using Smart Card and Train Log Data

As the mode share of the subway in Seoul has increased, the estimation of passenger travel routes has become a crucial issue to identify the congestion sections in the subway network. This paper aims to estimate the travel train of subway passengers in Seoul. The alternative routes are generated based on the train log data. The travel route is then estimated by the empirical cumulative distribution functions (ECDFs) of access time, egress time, and transfer time. The train choice probability is estimated for alternative train combinations and the train combination with the highest probability is assigned to the subway passenger. The estimated result is validated using the transfer gate data which are recorded on private subway lines. The result showed that the accuracy of the estimated travel train is shown to be 95.6%. The choice ratios for no-transfer, one-transfer, two-transfer, three-transfer, and four-transfer trips are estimated to be 53.9%, 37.7%, 6.5%, 1.5%, and 0.4%, respectively. Regarding the practical application, the passenger kilometers by lines are estimated with the travel route estimation of the whole network. As results of the passenger kilometer calculation, the passenger kilometer of the proposed algorithm is estimated to be 88,314 million passenger kilometer. The proposed algorithm estimates the passenger kilometer about 13% higher than the shortest path algorithm. This result implies that the passengers do not always prefer the shortest path and detour about 13% for their convenience.

Mineral lithotype identification on the andrill AND-2A drillcore, antarctica by using ternary mineral rock physics templates built from a self-consistent approach

In this work, wet bulk density ?WBD and compressional wave velocity VP core log data obtained along the AND-2A drillcore are plotted on density-velocity ternary mineral Rock Physics Templates (RPTs) built from a Self-Consistent (SC) micromechanics modelling with the purpose to deter- mine data trends that allow us to assist in identifying mineral lithotypes and lithological features throughout the 1138 m length of the drillcore. The elastic properties of the three dominant miner- als present in the drillcore (mixed clays, quartz, and calcite) and the pore-filling fluid (brine) were used as input data for the SC model. The interpreted lithology is then compared to that obtained from the analysis of the AND-2A drillcore ?WBD and VP log data using Gardner type density-velocity cross plots. Results from both the SC and Gardner methods are in good agreement with the main lithologies present in the AND-2A drillcore already reported in the scientific literature. Our findings also agree well when compared to the lithological description of six selected rock samples obtained at different depths on the AND-2A drillcore. These results suggest that the proposed SC approach could be helpful to assist to identify lithology in scientific drill holes where downhole elastic proper- ties may exist over intervals where portions of the drillcore were not recovered. Furthermore, even when elastic property data sets come from measurements on cores, the SC approach is likewise useful because, from visual analysis alone, lithology can sometimes be difficult to determine, and additional information from the analysis of the elastic properties may provide more insight.

THE APPLICATION OF THE SEMIPARAMETRIC GSTAR MODEL IN DETERMINING GAMMA-RAY LOG DATA ON SOIL LAYERS

This research examines the semiparametric Generalized Space-Time Autoregressive (GSTAR) spacetime modeling and determines its spatial weight. In general, the spatial weights used are uniform, binary weights, and based on the distance, the result is a fixed weight. The GSTAR model is a stochastic model that takes into account its random variables. Thus, it is necessary to study the random spatial weights. This study introduced a new method to estimate the observed value of the GSTAR model semiparametric with a uniform kernel. The data involved the Gamma Ray (GR) log data on four coal drill holes. The semiparametric GSTAR modeling aimed to predict the amount of log GR in the unobserved soil layer based on the observation data information on the layer above it and its surrounding location. The results revealed that semiparametric GSTAR modeling could predict the presence of coal seams and their thickness of drill holes. The results also highlight the validity test on the out-sample data that the error in each borehole results in a small error. In addition, the error tends to approach the actual observed value at a depth of 1 meter down.

Early Dropout Prediction in Online Learning of University using Machine Learning

Recently, most universities plan to open or open online learning courses, but the problem of dropout of online learning is still a problem for universities. Online learning has the advantage of being able to receive education anytime, anywhere, but it is true that the dropout rate is higher than offline classes because you have to manage and control your own study time without the help of a professor or manager. Therefore, it is very important for professors and managers to support students in a timely act to avoid the risk of dropout of university online classes. This study used the access log data recorded in the Learning Management System (LMS) and the learner's statistical information and calculated data, and aims to present predictive algorithms suitable for online learning dropout early prediction systems at universities. This study features a 7-year online learning history log data recorded in the Cyber University LMS system to overcome the data count limitations of existing studies and predict the risk of drop-out during the learning period. The characteristics of the data you utilized were used to validate the availability of predictive models by applying learner statistical information, number of system connections, number of lectures, previous semester grade data, machine learning based decision tree, arbitrary forest (RF), support vector machine (SVM) and deep learning (DNN). Studies show that random forest (RF) algorithms have the best prediction and performance, and deep learning algorithms also apply to learning management (LMS) systems.

Can System Log Data Enhance the Performance of Credit Scoring?—Evidence from an Internet Bank in Korea

The credit scoring model is one of the most important decision-making tools for the sustainability of banking systems. This study is the first to examine whether it can be improved by using system log data that are stoed extensively for system operation. We used the log data recorded by the mobile application system of KakaoBank, a leading internet bank used by more than 14 million people in Korea. After generating candidate variables from KakaoBank’s log data, we created a credit scoring model by utilizing variables with high information values and logistic regression, the most common method for developing credit scoring models in financial institutions. To prove our hypothesis on the improvement of credit scoring model performance, we performed an independent sample t-test using the simulation results of repeated model development and performance measurement based on randomly sampled data. Consequently, the discrimination power of the proposed model using logistic regression (neural network) compared to the credit bureau-based model significantly improved by 1.84 (2.22) percentage points based on the Kolmogorov–Smirnov statistics. The results of this study suggest that a bank can utilize the accumulated log data inside the bank to improve decision-making systems, including credit scoring, at a low cost.