Historical Logs Research Articles

Background Similarities as a Way to Predict Students’ Behaviour

The number of students opting for online educational platforms has been on the rise in recent years. Despite the upsurge, student retention is still a challenging task, with some students recording low-performance margins on online courses. This paper aims to predict students’ performance and behaviour based on their online activities on an e-learning platform. The paper will focus on the data logging history and utilise the learning management system (LMS) data set that is available on the Sakai platform. The data obtained from the LMS will be classified based on students’ learning styles in the e-learning environment. This classification will help students, teachers, and other stakeholders to engage early with students who are more likely to excel in selected topics. Therefore, clustering students based on their cognitive styles and overall performance will enable better adaption of the learning materials to their learning styles. The model-building steps include data preprocessing, parameter optimisation, and attribute selection procedures.

Development of the Brimmond Sand Fairway

Abstract The Eocene age Brimmond Sand Fairway is situated along the north-eastern flank of the Paleocene Forties Field (UKCNS blocks 21/10 and 22/6). Located along the western margin of this Brimmond Fairway are well imaged remobilized sands that form the reservoir interval for the Maule and Tonto Fields and, along with deep-water channels, the Brimmond Field. These Eocene Brimmond sandstones are encased in the Horda Shale which provides the sealing lithology. The interpretation of these remobilized and injected sands is driven from geometries derived from 3D seismic and historic logging of thin sandstones in the Eocene interval. Conical shape features with sills and steep dykes are mapped, with seismic evidence of injection along active faults and fractures. The developments of the Brimmond, Maule and Tonto Fields has been successful due to impressive seismic imaging with inversion and Direct Hydrocarbon Indicator (DHI) volumes allowing the identification of hydrocarbon bearing remobilized sandstones, along with 4D data imaging un-swept areas.

IDENTIFYING CARE ACTIVITIES THAT WERE SUPPORTED BY AMAZON ECHO FOR CARE PARTNERS AFTER ONSET OF COGNITIVE IMPAIRMENT

There are several important challenges when addressing the needs of older adults with cognitive impairment and their care partners including the potential for diminishing emotional well-being and loss of autonomy, which could potentially lead to a lower overall quality of life for both care partners (CPs). The motivation of this study was to identify the care activities that were supported by home based technology for care partners after the onset of cognitive impairment. This work was done through gathering multiple sources of qualitative and quantitative data, including mobile application dialogue history logs, pre and post interviews, user feedback groups and home visits. The technology deployed in the home of the care partners was a Voice User Interface Intelligent Agent, specifically the Amazon Echo with its intelligent agent “Alexa.” This technology was selected because it was not built from a traditional care model, yet embodies functions that could be used for all potential forms of care, including those that achieve a higher level of quality of life goals for care partners. From this study, we can further our understanding of how to deploy and design technology that shifts the perspective from “cure to care” with a focus on the older person and their lived experience, monitoring wellness, and not just addressing illness. Results and findings indicated that daily care activities of dyads that are seemingly fundamental are actually complex care activities that emerge from using the technology that support the care partners on multiple levels in satisfying multiple needs.

Research of Route Discovery using Proficient Route Encounter Algorithm in MANET

The objective of the research work is proposed an efficient node to node route discovery in Mobile AD HOC network based on neighboring nodes resent route discovery. MANET is always depends on selected and constant path with extended time period subsequently, and the flexibility of the battery power will be condensed in searching end nodes which roots numerous path failure. In most of the scenario path failure origins consistent route discovery which is normally reduces the efficiency and it will increase the dependency in computational overhead of the nodes in the mobile routing environment .In order to overcome such failure in MANET it is proposed to implementing route discovery methodologiesand it will be done through neighboring nodes with its resent history log which is an efficient factor to reduce delay between nodes and improve consistency comparing with existing system. DSR were experimentally implemented and compared to find the end to end delay, throughput and packet delivery ratio between nodes. The proposed work is focused towards effective node discovery through Proficient Route Encounter (PRE) algorithm which minimize the delay and energy resource of the path finding node, time taken to reach its requested node and maximize packet delivery ratio and its energy efficiency.

Smart Vehicle Monitoring and Analysis System with IOT Technology

In order to reduce the increasing number of road accidents in recent years, the Transport Ministry of Malaysia is keen to study the flexible road tax payment system based on the motorists’ driving behavior which can be monitored using a vehicle monitoring system. However, existing vehicle monitoring systems in the market mostly rely only on Global Positioning System (GPS) to estimate the vehicle’s speed and location which results may be inaccurate once the satellite coverage is lost due to tall buildings or dense forest. Therefore, a Smart Vehicle Monitoring and Analysis System (VMAS) using Internet of Things (IoT) technology is proposed. The developed system comprises of an Android application on smartphone, an On-board Diagnostics II (OBD-II) device and a cloud database management system. OBD-II device will collect real-time data on engine parameters. The data will be automatically extracted using Bluetooth connection to the Android platform for display and to upload data to the cloud server. The cloud server is used for data storage and can be accessed remotely by authorized or administrative personnel via a monitoring webpage. It will provide complete visibility of uploaded data in real time as well as instant activities reports and history logs. The performance investigation shows that the system latency is around 1 second. The data analysis includes driving attributes ranking and fuel efficiency rating which is foreseen to contribute significantly to flexible road tax scheme and diagnostic tests by vehicle manufacturers respectively.

Incremental Re-Identification by Cross-Direction and Cross-Ranking Adaption

Person re-identification is widely applied in video surveillance and criminal investigation applications. To achieve better performance, an additional re-ranking step is often exploited. Related methods attempt to optimize the result according to every single query independently . However, in a practical scene, as the investigation process goes on, the other queries, in particular, the gradually accumulated logs, can be used to guide or regularize the current query. In this paper, we propose to optimize the result according to not only the current query itself but also the other queries and historical logs. We respectively investigate the cross-direction and the cross-ranking constraints among different queries. Based on the investigations, we propose a reciprocal optimization method to refine multiple ranking lists reciprocally . Experiments on the VIPeR, new-protocol CUHK03, and Market-1501 datasets confirm the effectiveness of our method. In particular, on the Market-1501 dataset, with full utilization of the other queries, the method achieves an accuracy rate of 94.66% at rank-1 and a very high mAP of 75.12%, and significantly outperforms the state-of-the-art methods.

Structure and diversity of the Araucaria forest in southern Brazil: biotic homogenisation hinders the recognition of floristic assemblages related to altitude

Forest structure and species abundance may respond to environmental features triggered by geomorphology, altitude gradients and human impacts. In this study we investigated if data from a large-area inventory would support the segregation of floristic assemblages related to altitude-oriented forest physiognomies established for the Brazilian Araucaria forest, even after decades of intense logging and degradation by canopy opening, burning and cattle grazing/trampling inside forests. We applied cluster analysis and non-metric multidimensional scaling (NMDS) to group 143 systematically distributed 0.4 ha plots, located in the southern Brazilian Araucaria forest, according to their species abundance. Subsequently, forest structure attributes, species richness and diversity were assessed. We found 368 species and an average of ∼34 species per sample plot. Our estimations yielded a mean basal area per sample plot of 24.4 ± 2.1 m² ha−1, and a mean aboveground biomass of 86.9 ± 7.3 Mg ha−1 (± 95% confidence interval). Besides a subtle differentiation of the sample plots’ species abundance along the altitudinal gradient (514–1 560 m above sea level) revealed by the NMDS, we did not find ecologically meaningful floristic assemblages related to altitude. Possibly, this result is a consequence of floristic and stand structure homogenisation caused by intensive historical (and ongoing) logging, cattle grazing inside stands and general land-use changes. In turn, forest structure attributes, such as total tree height, tree density and basal area, were more evidently related to the altitude gradient, most likely due to lower temperatures, stronger winds, greater cloud cover and frost frequency occurring at high-altitude sites.

Sustainable maintainability management practices for offshore assets: A data-driven decision strategy

Efficient and reliable maintainability management is a basic element behind ensuring sustainable practices for offshore assets. A method for easily implementing offshore asset maintainability management is proposed and described in terms of the maintainability decision-making process based on data-mining technology. A dataset of historical logs from operational systems drawn from offshore oil and gas service firms that have adopted sustainable maintenance management information system practices, including 12 main systems and 91 auxiliary systems, is used to inform a case study describing maintainability management and the decision-making process. The method uses the optimal values from a multicore classification model that adopts a goal-oriented data mining decision-making approach. Based on maintenance feature attributes, fault duration, fault loss, and frequency of occurrence in maintenance management are the three most important predictors of decision objectives. The decision tree classification results also indicate that the total average maintainability of assets in the key assets component is 53.4%, and the total average maintainability of noncritical assets is 37.9%. The five most important characteristic events found during maintenance and configuration processes were flaws in the tubing for A-annulus communication, leakage in the closed position, external leakage, failure to close on demand, and hydraulic failures that cause safety loss. The results provide unique insights into how offshore enterprise operators can improve maintainability management and decision-making performance using a data-driven decision strategy perspective. Furthermore, it provides a solution for visual proactive maintenance management and decision making under a data-driven framework, making it easier to implement maintenance management and decision-making tasks.

A vocabulary recommendation method for spatiotemporal data discovery based on Bayesian network and ontologies

ABSTRACTIn the research field of spatiotemporal data discovery, how to utilize the semantic characteristics of spatiotemporal datasets is an important topic. This paper presented a content-based recommendation method, and applied Bayesian networks and ontologies into the vocabulary recommendation process for spatiotemporal data discovery. The source data of this research was from the MUDROD (Mining and Utilizing Dataset Relevancy from Oceanographic Datasets) search platform. From the historical search log, major keywords were extracted and organized according to ontologies in a hierarchical structure. Using the search history, the posterior probability between each subclass and their super class in the ontologies was calculated, indicating a recommendation likelihood. We created a Bayesian network model for inference based on ontologies. This model can address the following two objectives: (1) Given one class in the ontology, the model can judge which class has the biggest likelihood to be selected for recommendation. 2) Based on the search history of a user, the Bayesian network model can judge which class has the biggest probability to be recommended. Comparison experimentation with existing system and evaluation experimentation with expert knowledge show that this method is specifically helpful for spatiotemporal data discovery.

Innovative data regression incorporating deterministic knowledge for soft sensing in the process industry

Soft sensing is a monitoring technique for the indirect assessment of a target variable by means of direct measurements of others and the application of data mining on the historical log, as well as simplified models of the system. Due to technical and economic advantages respect to hardware sensing, soft sensing has been increasingly used in many scenarios, in particular within the process industry. Despite the literature being wide regarding the application of conventional regression techniques on data provided by the monitoring hardware, a systematic approach for supporting and improving data regression through the deterministic knowledge of the process is still missing. This contribution presents an innovative regression method based on first principle modeling. The method is introduced and tested under the case scenario of conventional clinker making process, for predicting the instantaneous fraction of the unburned calcium oxide. A second test case involves an innovative clinkering stage based on microwaves application.

Disturbance history mediates climate change effects on subtropical forest biomass and dynamics.

The responses of forest communities to interacting anthropogenic disturbances like climate change and logging are poorly known. Subtropical forests have been heavily modified by humans and their response to climate change is poorly understood. We investigated the 9‐year change observed in a mixed conifer‐hardwood Atlantic forest mosaic that included both mature and selectively logged forest patches in subtropical South America. We used demographic monitoring data within 10 1 ha plots that were subjected to distinct management histories (plots logged until 1955, until 1987, and unlogged) to test the hypothesis that climate change affected forest structure and dynamics differentially depending on past disturbances. We determined the functional group of all species based on life‐history affinities as well as many functional traits like leaf size, specific leaf area, wood density, total height, stem slenderness, and seed size data for the 66 most abundant species. Analysis of climate data revealed that minimum temperatures and rainfall have been increasing in the last few decades of the 20th century. Floristic composition differed mainly with logging history categories, with only minor change over the nine annual census intervals. Aboveground biomass increased in all plots, but increases were higher in mature unlogged forests, which showed signs of forest growth associated with increased CO2, temperature, and rainfall/treefall gap disturbance at the same time. Logged forests showed arrested succession as indicated by reduced abundances of Pioneers and biomass‐accumulators like Large Seeded Pioneers and Araucaria, as well as reduced functional diversity. Management actions aimed at creating regeneration opportunities for long‐lived pioneers are needed to restore community functional diversity, and ecosystem services such as increased aboveground biomass accumulation. We conclude that the effects of climate drivers on the dynamics of Brazilian mixed Atlantic forests vary with land‐use legacies, and can differ importantly from the ones prevalent in better known tropical forests.

A fuzzy logic approach to influence maximization in social networks

Within a community, social relationships are paramount to profile individuals’ conduct. For instance, an individual within a social network might be compelled to embrace a behaviour that his/her companion has recently adopted. Such social attitude is labelled social influence, which assesses the extent by which an individual’s social neighbourhood adopt that individual’s behaviour. We suggest an original approach to influence maximization using a fuzzy-logic based model, which combines influence-weights associated with historical logs of the social network users, and their favourable location in the network. Our approach uses a two-phases process to maximise influence diffusion. First, we harness the complexity of the problem by partitioning the network into significantly-enriched community-structures, which we then use as modules to locate the most influential nodes across the entire network. These key users are determined relatively to a fuzzy-logic based technique that identifies the most influential users, out of which the seed-set candidates to diffuse a behaviour or an innovation are extracted following the allocated budget for the influence campaign. This way to deal with influence propagation in social networks, is different from previous models, which do not compare structural and behavioural attributes among members of the network. The performance results show the validity of the proposed partitioning-approach of a social network into communities, and its contribution to “activate” a higher number of nodes overall. Our experimental study involves both empirical and real contemporary social-networks, whereby a smaller seed set of key users, is shown to scale influence to the high-end compared to some renowned techniques, which employ a larger seed set of key users and yet they influence less nodes in the social network.

Fire and Forest Management in Montane Forests of the Northwestern States and California, USA

We reviewed forest management in the mountainous regions of several northwestern states and California in the United States and how it has impacted current issues facing these forests. We focused on the large-scale activities like fire suppression and logging which resulted in landscape level changes. We divided the region into two main forests types; wet, like the forests in the Pacific Northwest, and dry, like the forests in the Sierra Nevada and Cascade ranges. In the wet forests, the history of intensive logging shaped the current forest structure, while fire suppression played a more major role in the dry forests. Next, we looked at how historical management has influenced new forest management challenges, like catastrophic fires, decreased heterogeneity, and climate change. We then synthesized what current management actions are performed to address these issues, like thinning to reduce fuels or improve structural heterogeneity, and restoration after large-scale disturbances. Lastly, we touch on some major policies that have influenced changes in management. We note a trend towards ecosystem management that considers a forest’s historical disturbance regime. With expected climate induced changes in fire frequency, it is suggested that fuel treatments be implemented in dry forests to ensure an understory fire regime is restored in these forest systems. With respect to wet forests in this region, it is suggested that there is still a place for stand-replacing fire regimes. However, these forests will require structural changes incorporating heterogeneity to improve their resiliency and health.

IPACS: a physical access control system as a service and mobile application

We present iPACS, an automated physical access control system as a cloud service for controlling users activity, navigation and access in large residential infrastructures (e.g. apartment buildings, shopping malls). The main idea is to install on buildings, beacon Bluetooth radio transmitters that broadcast their identifier to nearby devices (e.g. users’ mobile phones) and from there, transmit this information safely, together with other application and user data, to a private cloud. The service tracks people, monitors the overall activity in buildings and public venues and offers prompt response in cases of critical events (i.e. overcrowding, health incidents, attacks etc.). iPACS provides also services aiming to inform the infrastructure manager for possible increase of users activity or access requests that require permission based on subscriptions or authorization criteria. These services are deployed over secure private clouds capable of dealing safely with sensitive information while ensuring users privacy. Collectively, anonymous history (log) data are sent to a public cloud for analysis.

Main ecological drivers of woody plant species richness recovery in secondary forests in China

Identifying drivers behind biodiversity recovery is critical to promote efficient ecological restoration. Yet to date, for secondary forests in China there is a considerable uncertainty concerning the ecological drivers that affect plant diversity recovery. Following up on a previous published meta-analysis on the patterns of species recovery across the country, here we further incorporate data on the logging history, climate, forest landscape and forest attribute to conduct a nationwide analysis of the main drivers influencing the recovery of woody plant species richness in secondary forests. Results showed that regional species pool exerted a positive effect on the recovery ratio of species richness and this effect was stronger in selective cutting forests than that in clear cutting forests. We also found that temperature had a negative effect, and the shape complexity of forest patches as well as the percentage of forest cover in the landscape had positive effects on the recovery ratio of species richness. Our study provides basic information on recovery and resilience analyses of secondary forests in China.

Fire and legacy effects of logging on understorey assemblages in wet-sclerophyll forests

Wet sclerophyll forests of south-eastern Australia typically experience wildfire once or twice a century. However, disturbance regimes have changed drastically in recent decades due to clear-fell logging and altered fire regimes. To date, botanical research on disturbances in wet-forests has focussed on individual elements of disturbance regimes, such as intensity, at single points in time, largely neglecting past disturbance history. Studies of the impact of previous disturbance history on plant responses to successive disturbance events are important to our understanding of vegetation dynamics. Here we investigate the response of wet-forest understorey species to two important elements of disturbance regimes – timing and type – and trajectories of change in these vegetation communities. In surveys separated by 15 years over 128 sites, we recorded the frequency of occurrence of 21 understorey species from stands with disturbance histories ranging from 4 years post clear-fell logging to 150 years post wildfire. Approximately half our sites were burnt in the 2009 Black Saturday wildfires. This provided an opportunity to examine the effects of inter-fire interval and the legacy effects of clear-fell logging. Generalised linear mixed modelling showed that many of the species studied appear to be at risk of population decline as a result of clear-fell logging. Unlike wildfire, clear-fell logging led to changes in the understorey, having a long-lasting impact on the presence of key wet-forest taxa that rely on vegetative regeneration. These include large shade-bearing shrubs like Hedycarya angustifolia R.Cunn., Bedfordia arborescens Hochr. and Olearia argophylla (Labill.) Benth., which were resilient to recurrent wildfire but responded negatively to recent wildfire in sites with a history of clear-fell logging. Negative effects of a short inter-fire interval were limited to a few species.

Heuristic rule-based process discovery approach from events data

Knowledge management consists of transforming data into beneficial knowledge in a business environment. Today, large amounts of data related to the execution of business processes called event logs are stored in the information systems. Process mining enables knowledge management by extracting knowledge from these historical event logs. Most organisations seek to understand how their business processes are executed to improve them. Therefore, several process discovery techniques have been developed in the field of process mining. However, none of the existing algorithms can discover all types of process constructs that can exist in an event log in a restricted time. This paper proposes a new heuristic rule-based technique that is capable of constructing process models with standard constructs, short loops, invisible tasks, duplicate tasks, and non-free choice constructs. Artificial and real-life data have been used to evaluate the algorithm. The results demonstrate that the aforementioned characteristics can be discovered correctly.

The Role of the User's Browsing and Query History for Improving MPC-generated Query Suggestions

In this paper we use the user's recent web browsing history in order to provide better query suggestions in an information retrieval system. We have built a Chrome browser plugin that collects each web page visited by a user and submits it to our query suggestion server for indexing, thus building a personal history profile for each user. We then analyze if future queries submitted by a user to the search engine can be predicted from web pages visited by that user in the past (i.e. his recent browsing history) or from queries submitted by that user in the past (i.e. his recent query history). The contribution of this paper is a method of using this personal history profile for reordering the query suggestions offered by Google when the user searches information on Google, moving query suggestions more relevant to the user's information need to the front positions in the Google provided query suggestions list. We have collected browsing history log data for over 4 months from several users who installed our Chrome plugin on their local computers and then we performed an offline evaluation test that shows that our personalized query suggestion system improves the MRR (i.e. Mean Reciprocal Rank) score of Google query suggestions by approximately 0.04 (i.e. improves Google's MRR score by 12 percents).

An Estimate of the Financial Cost of Peatland Restoration in Indonesia

The peat forests of Indonesia have experienced extensive deforestation and degradation over recent decades. High demand for Indonesian timber and plantation development has driven large-scale draining and clearing of peat forest, resulting in extensive fires and smoke haze problems across the region. These fires caused more than 100,000 premature deaths in 2015 alone, increased the pressure on several already threatened species, and placed Indonesia among the top greenhouse gas emitting countries globally. In response, the Indonesian government has launched an initiative to restore more than 2 million ha of peatland between now and 2020. Although there is a substantial body of academic literature that deals with technical aspects of tropical peatland restoration, little is published on the costs of tropical peatland restoration activities. In this study, we examine the case of peatland restoration in the provinces of Kalimantan, Sumatra, and Papua in Indonesia, and propose a restoration activity classification scheme based on fire, drainage, and logging history of peatland areas. We use this scheme to identify the restoration activity needs of different areas and then develop a preliminary gross financial cost estimate for the restoration activities proposed under the national 2-million-ha peatland restoration initiative. We find that it is likely to cost more than US$4.6 billion to complete the national 2-million-ha restoration initiative, which is substantially more than the funds currently allocated to the challenge across Indonesian and international donor budgets.

DeepType

Mobile users spend an extensive amount of time on typing. A more efficient text input instrument brings a significant enhancement of user experience. Deep learning techniques have been recently applied to suggesting the next words of input, but to achieve more accurate predictions, these models should be customized for individual users. Personalization is often at the expense of privacy concerns. Existing solutions require users to upload the historical logs of their input text to the cloud so that a deep learning predictor can be trained. In this work, we propose a novel approach, called DeepType, to personalize text input with better privacy. The basic idea is intuitive: training deep learning predictors on the device instead of on the cloud, so that the model makes personalized and private data never leaves the device to externals. With DeepType, a global model is first trained on the cloud using massive public corpora, and our personalization is done by incrementally customizing the global model with data on individual devices. We further propose a set of techniques that effectively reduce the computation cost of training deep learning models on mobile devices at the cost of negligible accuracy loss. Experiments using real-world text input from millions of users demonstrate that DeepType significantly improves the input efficiency for individual users, and its incurred computation and energy costs are within the performance and battery restrictions of typical COTS mobile devices.