Predictive Modeling

Keywords: MIMO antenna UWB antenna Model predictive Control Non-uniform microstrip line Printed monopole antenna a b s t r a c t A novel ultra wideband (UWB) printed monopole multiple-input multiple-output (MIMO) antenna with non-uniform transmission line using nonlinear model predictive control (NMPC) is presented. The proposed antenna is superior to conventional antennas in terms of dimensions, gain, and efficiency while maintaining the impedance bandwidth. In order to improve the results, a non-uniform transmission line has been used for impedance matching between the radiated patch element and the coaxial cable. For designing the non-uniform transmission line, it has been expanded using cosine terms. Regarding the presence of differential equation for the variation in the impedance of the transmission line and its transformation to the state-space equation, NMPC has been employed to design the transmission line and determine the cosine expansion coefficients. Two base antennas, as MIMO, were simulated configuration and fabricated. The surface area of the proposed MIMO antenna is 0.99 k 2 g , the wavelength has been obtained for the center frequency of the 3.16 GHz to 10.6 GHz range, and its mutual coupling, peak gain, channel capacity loss (CCL), total active reflection coefficient (TARC), mean effective gain (MEG) and diversity gain (DG), envelope correlation (ECC) are acceptable. The simulation and measurement results are in good agreement, and the proposed antenna is suitable for MIMO applications. Ó 2020 The ''Authors". Karabuk University. Publishing services by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

This paper discusses territorial markers in the Late Mesolithic based on ethnoarchaeological data as well as findings from modern social psychology. On the basis of the archaeological evidence from Southern Scandinavia, it is suggested that Late Mesolithic graves, which often appear to have had visual markers above ground and probably, in some instances, were ‘air graves’ located in trees or on platforms, were used as the primary territorial markers, whereas coeval shell middens played a role as site markers at archaeological sites where molluscs were collected.

On the basis of the evidence from the mouth of the River Muge in Portugal, it is suggested that the shell middens here may have constituted the primary territorial markers.

In general, the picture that emerges is of aggressive, territorial and non-egalitarian Late Mesolithic societies, in good accordance with the skeletal record .

The Sunda clouded leopard Neofelis diardi is a medium sized (15–25 kg) cat, found only on the Sundaic islands of Borneo and Sumatra. In recent years intensive camera-trapping surveys in Borneo have begun to shed light on the habitat associations and basic ecology of this elusive wild cat, but its distribution on an island-wide scale remains very poorly known. Such information is an essential element in the assessment of the Sunda clouded leopard’s conservation status and in the development of conservation action. In this paper we use MaxEnt niche distribution modelling to make predictions regarding the current distribution of this cat on Borneo. We collected a total of 259 occurrence records for the Sunda clouded leopard, stemming from all Bornean regions apart from Brunei and South Kalimantan, of which 48 (Balanced Model) or 94 (Spatial Filtering Model) were used in our modelling. Our habitat suitability model suggests that this species has a widespread distribution over a large co...

The Early Mesolithic of England has been studied in detail over the past 50 years, and a large amount of important literature has been produced on the site of Star Carr and seasonality, but a predictive model for this period has never been attempted.  Site discovery for this period thus relies upon finds of flint scatters.  Predictive modelling has been used to describe site appearance, and has shown some success in the Netherlands, and America as a cost efficient alternative to individual desk based assessments of sites.  Predictive models have also been used as a tool in archaeological research in the past.  My study aimed to create a predictive model which could be used to prospect for archaeological sites and to discuss various issues around the Mesolithic sites seen here.  There will be a discussion of scale within analyses of the Mesolithic in England.  Further, the reasons for the negative results from this process will have to be discussed.  My study has been attempted using open source GIS software (GRASS and/or QGIS), and it is my hope that it can be made available as a GIS resource.

The accuracy of a predictive system is critical for predictive maintenance and to support the right decisions at the right times. Statistical models, such as ARIMA and SARIMA, are unable to describe the stochastic nature of the data. Neural networks, such as long short-term memory (LSTM) and the gated recurrent unit (GRU), are good predictors for univariate and multivariate data. The present paper describes a case study where the performances of long short-term memory and gated recurrent units are compared, based on different hyperparameters. In general, gated recurrent units exhibit better performance, based on a case study on pulp paper presses. The final result demonstrates that, to maximize the equipment availability, gated recurrent units, as demonstrated in the paper, are the best options.

The Boston Police Department has limited resources and cannot easily patrol and guard the entire area under its jurisdiction. This situation limits its immediate intervention capacity to the patrolled area: forecasting dangerous criminal misconduct in a urban area throughout data analytics techniques would enable Boston Police to have better visibility on future potential crimes. For this project we are going to focus on predicting time and location of  criminal offences, with the ultimate goal to help the Boston Police Department with a model that offer meaningful insights on the crime patterns in the city.
Our data includes time, location and type of crime, weather conditions at the time and socioeconomic data of the area where it happened: these are all sourced from public entities such as the City of Boston portal, the NOAA and the US Census. To simplify the original dataset we reduced the number of possible locations by splitting Boston into a grid, defining ~ 500 areas; secondly, we summarized the 30+ different crime types into violent crimes, property crimes and other crimes, and we computed a series of time series variables accounting for the frequency of crime in each area in the previous 24 hours, week, one month and four months.
Because of the inherently different causes that might determine a violent versus a crime against property, we decided to have two separate models for forecasting the two types of crimes. In both cases, given the binary nature of the dependent variable (the occurrence of violent or property crimes for a given date and location) we developed a logistic regression and a random forest model. Specifically, as imaginable given the variable transformations described above, the model is intended to be a mixed of time series features, and “fundamentals” (income), similarly to what is done with financial models. We found that areas that have been experiencing violent crimes in the past are more likely to be affected by the same problem tomorrow. Also, the colder and the rainer outside, the less violent crimes there will be on the streets of Boston. Finally, since the average income is positively correlated and the median income is negatively correlated with the probability of both property and violent crime, this suggests that - as intuitively true - the difference between the two - representing income inequality - cause higher level of crime.
Our logistic model succeeds in delivering a basic forecasting tool that would allow Boston Police to be present - and therefore possibly deter and prevent - in 71.1% of areas in which a violent crime is happening, by patrolling only 30% of the city. Similarly, by patrolling 30% of the town, the Police would be able to be in the same area where a property crime is happening 64.9% of the times. We also tried fitting a Random Forest model on the data, but it yielded poorer results. Besides of what percentage of the city of Boston the Police is able to patrol, we can effectively interpret the ROCR curve as an optimization tool: given limited resources, the police department should give priority to patrol the locations  featuring the highest probability of crime happening in a given day. In this sense our model gives the law enforcement agencies an edge against micro criminality in the city of Boston as well as a tool to counterbalance possible budget cuts by optimizing Police presence on the ground.

Educational data mining (EDM) can be used in extracting useful patterns in students' academic records which may aid in management decision making on determination of engineering students' specialization track once the general engineering academic requirements were completed. The objective of the research is to provide a specialization selection recommendation for engineering students through application of data mining algorithm and adoption of the rule sets generated by a predictive model. The attributes that may be significant in creating prediction were determined using correlation-based feature selection. Comparative analysis among known algorithms shows that the highest accuracy was considered. A decision tree classification model using WEKA and J48 produced an accuracy value of 80.06. The study revealed that Gender, Algebra, Calculus and physics courses found to have significant effect in predicting the engineering specialization, thus strengthening the general notion that for engineering students to be more successful, the academic performance with the above courses should be highly considered.

Direct mail marketing will continue to be successful when tools are used to optimize audience selection. Audience selection for direct mail campaigns should be smarter and more personalized in the growing presence of online marketing. Road Scholar has been offering learning adventures in all 50 states and 150 countries worldwide to adults aged 50+ since 1975. Direct mail has been and continues to be an important channel to Road Scholar due to the older audience, but as a not-for-profit we need to use our budget wisely. Historically, audience selections for our catalog campaigns relied on limited segmentation and were not targeted. In 2015, we implemented advanced predictive modeling as an additional tool to the audience selection process via JMP Pro. We developed, validated, and pre-assessed predictive response models by using JMP Pro to identify key independent variables predictive of the desired response using historical data. Post-campaign analyses show that advanced predictive modeling identified high performing customers in the top deciles. As a result, we have been able to take a more targeted approach while still increasing response. The results demonstrate that direct mail campaigns are still effective when advanced tools are used to optimize audience selection.

Carbon dioxide (CO2) emissions have been continuously escalating in recent years. The escalating trend is consistent with the current economic activities and other uncertain variables such as demand and supply in businesses and energy needs. Linear model is one of the most commonly used methods to explain the relationship between CO2 emissions and the related economic variables. The conventional linear regression model has a disadvantage in describing the relationships due to the variables' uncertainty and vague information. To address this problem, the fuzzy linear regression model has been proposed for explaining the relationships. However, the performance of the two linear models for predicting CO2 emissions is not immediately known. This paper presents a comparative study of conventional linear regression model and linear regression with fuzzy numbers model for predicting CO2 emissions in Malaysia. Twenty five years data from 1981 to 2005 of CO2 emissions, fuel mix, transportation, gross domestic product, and population have been used to develop the model of possibilistic fuzzy linear regression (PFLR) and multiple linear regression (MLR). The criteria of performance evaluation are calculated for estimating and comparing the performances of PFLR and MLR models. The performance comparison of PFLR and MLR models due to mean absolute percentage errors, root mean squared error criteria; indicate that MLR performed better on CO2 emissions prediction. A considerable further work needs to be done to determine the flexibility of fuzzy numbers in enhancing the performance of PFLR against the MLR.

الملتقى العلمي  الدولي 12-13 نوفمبر 2017 حول
التحول الرقمي للمؤسسات والنماذج التنبؤية
علىالمعطيات الكبيرة
Colloque international12-13 Nov 2007
La transformation numérique des entreprises & les modèles prédictifs sur Big Data
International Colloquial 12-13 Nov 2007
The digital transformation of enterprises & the predictive models on big data


=============
La co-construction de la confiance numérique selon la théorie des capacités d’absorption combinée avec le courant confiance et gestion
Pr Mokrane REFAA
البناءالمشتركللثقةالرقميةوفقالنظريةقدراتالاستيعاب مشتركةمع حركةالثقةوالتسيير
البروفسور مقران رفاع
The co-construction of digital trust according to the theory of absorption capabilities combined with the current trust and management
Pr Mokrane REFAA

—Load forecasting, particularly short-term load forecasting (STLF) plays a vital role in the economy streaming and tracking of power system. Many stochastic and artificial intelligence techniques haven been used in order to come up with an accurate (less error) short-term load forecast. Here, we introduce a new approach to short-term load forecasting (STLF) using the conventional Hidden Markov Model (HMM) then compare it with Autoregressive Integrated Moving Average (ARIMA) models. Three-dimensional continuous multivariate Gaussian emission probabilities are used in this experiment for HMM. Meanwhile for ARIMA models, different parameters are used for different kinds of dataset. Comparison is done afterwards to the actual load value using MAPE and RMSE.

Petroleum production and export play a dominant role in Nigeria's economy and account for about 90% of her gross earnings. This dominant role has pushed agriculture, the traditional mainstay of the economy, from the early fifties and sixties, to the background. In this research work we fitted a univariate Seasonal Autoregressive Integrated Moving Average model (SARIMA) to the monthly crude oil production in Nigeria between 2002 and 2016. Different Box-Jenskin Autoregressive Integrated Moving Average (ARIMA) models are fitted and diagnosed. However, ARIMA (2,1,0)(2,1,1) 12 was the best model for the data. The model was further validated and it was discovered that autocorrelation between residuals at different lag times was not significant. Finally, the time plot of the in-sample forecast errors shows that the variance of the forecast errors seems to be roughly constant over time and the histogram of the time series shows that the forecast errors are roughly normally distributed and the mean seems to be close to zero, it seems plausible that the forecast errors are normally distributed with mean zero and constant variance.

Mathematics plays a critical role in many modeling and simulation applications, and whenever detailed representations of combat phenomenon are required (e.g., attrition, detection, etc.). The required mathematics can seem daunting, but a wealth of problems relevant to the M&S community can be addressed with efficient numerical algorithms and heuristics. We shall present a small sampling of these methods and relevant examples of their usage. Examples include combat attrition, target engagement, detection, and missile defense. Most of these can be implemented with a spreadsheet for demonstration.

Breast cancer poses serious threat to the lives of people and it is the second leading cause of death in women today and the most common cancer in women in developing countries in Nigeria where there are no services in place to aid the early detection of breast cancer in Nigerian women. A number of studies have been undertaken in order to understand the prediction of breast cancer risks using data mining techniques. Hence, this study is focused at using two data mining techniques to predict breast cancer risks in Nigerian patients using the naïve bayes’ and the J48 decision trees algorithms. The performance of both classification techniques was evaluated in order to determine the most efficient and effective model. The J48 decision trees showed a higher accuracy with lower error rates compared to that of the naïve bayes’ method while the evaluation criteria proved the J48 decision trees to be a more effective and efficient classification techniques for the prediction of breast cancer risks among patients of the study location.

Borneo harbours more endemic carnivores than does any other island in the world except Madagascar, and almost half of the Bornean carnivore species have been classified by The IUCN Red List of Threatened Species as globally threatened. Here, a systematic conservation planning framework highlighted key carnivore landscapes, conservation research and intervention priorities, and gaps in current knowledge of Bornean carnivore ecology. All single-species predictive habitat suitability index (HSI) models presented in this issue (20 species, comprising all carnivores on Borneo except otters [Lutrinae] and sun bear Helarctos malayanus) were standardised by converting HSI values into binary maps, and combined to derive species richness maps to discuss and delineate areas of conservation priority. The highest predicted carnivore species richness (defined here as the sum of the binary threshold maps), corresponds to interior lowland, upland and lower montane forest, whereas areas with lowest predicted species richness correspond to coastal lowlands already largely converted to oil palm plantations. The 12 proposed areas of conservation importance for carnivores focus on large landscapes and connectivity between subunits, many centred around the tri-national Heart of Borneo initiative, with additional areas for wetland/lowland species. A large proportion of these proposed conservation landscapes are being exploited for commercial purposes (e.g., logging concessions) and would, therefore, improve in conservation value if their management became more sustainable. The most important research priorities for Bornean carnivores are species resilience to altered and fragmented landscapes; under-surveyed regions; and the effects and relative intensity of hunting across the island. The most pressing conservation interventions include conservation research on the most threatened Bornean carnivores: the Bornean ferret badger Melogale everetti and Hose's civet Diplogale hosei (highland endemics), and the flat-headed cat Prionailurus planiceps and otter civet Cynogale bennettii (wetland specialists). Targeted conservation research and integration of research findings into decision-making, maintaining and restoring connectivity, raising awareness and improving enforcement and governance are also important conservation interventions. Although more resources are needed for conservation and research, the joint effort of scientists, conservationists and government authorities in the identification of key carnivore landscapes, research priorities and conservation issues which this study presents raises hope that more targeted conservation efforts for Bornean carnivores will follow in the future.

Abstrak (Bahasa Indonesia). Wilayah Borneo memiliki paling banyak karnivora endemik dibandingkan pulau lainnya di seluruh dunia, selain Madagaskar. Hampir setengah dari keseluruhan jenis karnivora yang ada di Borneo masuk dalam kategori terancam pada daftar IUCN Red List of Threatened Species. Disini, suatu kerangka perencanaan konservasi yang sistimatis menyoroti kawasan kunci karnivora, prioritas penelitian dan intervensi aspek konservasi, serta kesenjangan dari pengetahuan mengenai ekologi karnivora di Borneo. Semua model pendugaan Index Kesesuaian Habitat (IKH) pada setiap spesies yang disajikan dalam tulisan ini (20 spesies, terdiri atas semua spesies karnivora yang ada di Borneo terkecuali berang-berang [Lutrinae] dan Beruang Madu Helarctos malayanus) distandardisasi dengan mengkonversi nilai IKH ke dalam pemetaan binari, dan digabung untuk memperoleh peta kekayaan spesies guna dibahas dan ditetapkan wilayah prioritasnya. Nilai pendugaan kekayaan spesies karnivora tertinggi (sebagai jumlah dari peta ambang binari), adalah selaras dengan bentuk hutan pedalaman dataran rendah, hutan dataran tinggi dan hutan pegunungan dataran rendah, sedangkan wilayah dengan pendugaan kekayaan spesies karnivora terendah adalah selaras dengan bentuk pantai dataran rendah yang telah berubah sepenuhnya menjadi perkebunan kelapa sawit. Dua belas wilayah dicanangkan sebagai wilayah penting untuk dikonservasi bagi karnivora, terpusat pada bentang alam yang luas dan saling berhubungan diantara sub-unit, kebanyakan berpusat di sekitar area Heart of Borneo di tiga negara, dengan wilayah tambahan untuk spesies dataran rendah atau lahan basah. Sebagain besar dari wilayah yang dicanangkan ini merupakan wilayah yang telah banyak dieksploatasi untuk tujuan komersil (konsesi hutan) dan dapat meningkatkan nilai konservasi manakala pengelolaannya bersifat berkelanjutan. Prioritas terpenting untuk penelitian karnivora Borneo adalah kesanggupan spesies beradaptasi pada wilayah yang terfragmentasi; wilayah-wilayah dalam area survey, serta pengaruh dari perburuan di semua wilayah pulau. Intervensi konservasi terpenting mencakup penelitian konservasi pada karnivora Borneo yang paling terancam: Biul Slentek Melogale everetti dan Musang Gunung Diplogale hosei (endemik dataran tinggi), serta Kucing Tandang Prionailurus planiceps dan Musang Air Cynogale bennettii (spesialis lahan basah). Penelitian wilayah konservasi terpilih dan integrasi temuan hasil penelitian kedalam penetapan kebijakan, menjaga dan memperbaiki RAFFLES BULLETIN OF ZOOLOGY Supplement

In this paper we aim to show that motor intentionality, as the underlying ground for social cognition, can be explained through the predictive engagement model. Sensorimotor processes seem to play central roles in social interaction, cognition and language. We question the phenomenologi-cal role of the body in social cognition and further investigate a causal neural explanation. We will adopt a different perspective by linking the role of the body and intercorporeality with recent findings in philosophy of neuroscience under the predictive brain hypothesis. In fact, the living body seems to entertain a dialogical and enactive relationship with the surrounding context, as well as with neural circuits actively responding to external stimuli. The body is thus configured as a living organism, and not as a mere biological substratum, offering to phenomenology and empirical sciences further confirmations of the possibility-and need-for a cooperation.

Remediation of the heavy organic deposits in the course of petroleum production has been a costly process and it has hampered the production activities in many parts of the world. In this report the authors study the in situ remediation of heavy organics including asphaltene, resin, paraffin I wax, etc. in oil production and transportation practices and its relation with the laboratory and theoretical simulation experiments. It is demonstrated that while asphaltene is the major compound responsible for various arterial blockage cases in the petroleum industry its interaction with other heavy organics is a major factor whether depositions will occur or not. Various mechanisms of asphaltene flocculation and micelle formation are reviewed and illustrative pathways are presented for each mechanism. The current in situ remediation and laboratory simulation practices in the oil industry to combat the problem is reviewed and analysis are made of the cause and effect of each factor influencing them. It is demonstrated that with the development of predictive modeling of heavy organic deposition one can make alterations in the production and transportation schemes to reduce and eliminate the deposition problem. With the development of predictive models one can also screen the candidate dispersants and solvents from the points of view of availability at the site, effectiveness and economy.

O objetivo deste trabalho foi desenvolver e comparar modelos preditivos para detecção de diabetes não diagnosticado utilizando diferentes algoritmos de aprendizagem de máquina. Os dados utilizados foram do Estudo Longitudinal de Saúde do Adulto (ELSA-Brasil), um conjunto bastante completo com aproximadamente 15 mil participantes. As variáveis preditoras foram selecionadas de forma que sejam informações simples dos participantes, sem necessidade de exames laboratoriais. Os testes foram realizados em quatro etapas: ajuste dos parâmetros através de validação cruzada, seleção automática de variáveis, validação cruzada para estimativa de erros e teste de generalização em um conjunto independente dos dados. Os resultados demonstram a viabilidade de utilizar informações simples para detectar casos diabetes não diagnosticado na população. Além disso, os resultados comparam algoritmos de aprendizagem de máquina e mostram a possibilidade de utilizar outros algoritmos, alternativamente à Regressão Logística, para a construção de modelos preditivos.

The goal of this study is to overcome the identified methodological limitations of prior studies aimed at predicting the type of auditor opinion and draw definite conclusions on the relative predictive performance of different predictive methods for this particular task. Predictive performance of twelve candidate models from the realms of statistics and machine learning is assessed separately for the two common real-life scenarios: a) when prior information on the client (i.e. types of audit opinion received in the past) is available and can be used in prediction, and b) when such information is not available (e.g. new companies). The results show that, in the first scenario, several methods from both realms achieve comparable predictive performance of around 0.89, as measured by the Area under the curve (AUC). In the second scenario, however, machine learning algorithms, particularly tree-based ones, such as random forest, perform significantly better, achieving the AUC of up to 0.79. Finally, we develop and assess the predictive performance of two hybrid models aimed at combining the strong points of both statistical (i.e. interpretability of results) and machine learning (i.e. handling a large number of predictors and improved accuracy) approaches. The complete procedure is demonstrated in a reproducible manner, using the largest empirical data set ever used in this stream of research, comprising 13,561 pairs of annual financial statements and the corresponding audit reports. The procedures described in this study allow audit and finance professionals around the globe to develop and test predictive models that will aid their procedures of audit planning and risk assessment.

Abstract
Purpose – The purpose of this paper is to examine the interrelationships among the extracted
constructs of airline service quality (SQ), perceived image, perceived value, passenger satisfaction and
their influence on passengers’ future behavioral intentions in the domestic aviation sector market in
Indian context.
Design/methodology/approach – A conceptual framework is developed based on previous studies.
In order to address a lack of comprehensive evaluation of airline SQ, this study extracted the
dimensions of airline SQ and used them in the structural model. The conceptual model is tested using
structural equation modeling approach using maximum likelihood estimation technique. A sample of
526 domestic full-service passengers was used to test the hypothesized relationships among
the constructs.
Findings – Out of the three dimensions of airline SQ extracted, convenience and promptness with
reliability dimension of SQ was having a direct influence on perceived image, perceived value and
passenger satisfaction which in turn influence BI. The findings of this study indicate the passenger
evaluation process flows from perceived image to satisfaction via passenger perceived value and not
directly from perceived image to passenger satisfaction. Only passengers’ satisfaction was found to
have a direct influence on passengers’ future BI.
Originality/value – This study modified SERVPERF instrument for airline settings and empirically
validated the instrument using data from Indian domestic passengers. This research adds to the body of
knowledge by providing a conceptual framework which gives an empirical insight into the interrelationship
among the constructs of airline SQ, perceived image, perceived value, passenger satisfaction and its effects
on passenger’s future BI. The conceptual model developed and tested in this paper will act as a reference in
forming suitable marketing strategies for airline’s competitive advantage.
Keywords India, Behavioural intentions, Airlines, Exploratory factor analysis (EFA),
Confirmatory factor analysis (CFA), Measurement model, Structural equation modelling (SEM),
Perceived image, Perceived value, Service quality, Customer satisfaction
Paper type Research paper

The flat-headed cat Prionailurus planiceps is classified as one of the most threatened cat species in the world. Its range is restricted to southern Thailand, peninsular Malaysia and the two largest Sunda Islands, Borneo and Sumatra. Its association with wetlands and lowland areas puts great pressure on this species, because these habitats are most threatened by degradation and agricultural conversions. Borneo has been identified as the stronghold for flat-headed cat. Of 140 flat-headed cat occurrence records for Borneo, 50 (Balanced Model) or 76 (Spatial Filtering Model) were used to estimate potential habitat suitability. Although we predicted suitable habitat for the flat-headed cat scattered across the lowlands of Borneo, some large lowland areas are predicted to be unsuitable, a likely consequence of forest conversion to oil palm plantations. Of particular predicted importance are forests in Brunei Darussalam, the Sabangau National Park and surrounding forest complex in Central Kalimantan and forests in North Kalimantan, as well as the central forest block in Sabah. The main threat to the flat-headed cat is ongoing transformation of forested areas to monoculture plantations, as the species appears unable to use these human-dominated habitats. Of particular importance for long-term survival of flat-headed cat is conservation of land near rivers and peat swamp forests.

Abstrak (Bahasa Indonesia). Kucing Tandang Prionailurus planiceps diklasifikasikan sebagai salah satu jenis kucing liar yang paling terancam di dunia. Daerah sebarannya terbatas di Thailand Selatan, Semenanjung Malaysia dan dua Kepulauan Sunda terbesar, Borneo dan Sumatera. Asosiasi jenis ini adalah lahan basah dan dataran rendah, yaitu kawasan yang mendapatkan tekanan yang sangat tinggi saat ini oleh degradasi dan konversi menjadi lahan pertanian, sehingga tekanan terhadap Kucing Tandang juga sangat tinggi. Borneo telah diidentifikasi sebagai benteng terbaik untuk Kucing Tandang. Kami mengumpulkan 140 catatan kehadiran Kucing Tandang di Borneo, dimana 50 (Model Penyeimbang) atau 76 (Model Spasial Tersaring) digunakan untuk mengestimasi habitat yang sesuai. Walaupun prediksi habitat yang sesuai tersebar di dataran rendah Borneo, tetapi hasil penelitian kami juga menyatakan sebagian besar kawasan dataran rendah tidak sesuai untuk habitat, yang merupakan konsekuensi dari konversi hutan menjadi perkebunan sawit. Prediksi kawasan penting adalah kawasan hutan di Brunei Darussalam, Taman Nasional Sebangau dan sekitar kawasan hutan di Kalimantan Tengah, kawasan hutan di Kalimantan Utara, juga kawasan hutan di Sabah. Ancaman utama jenis Kucing Tandang adalah konversi kawasan hutan menjadi kawasan tanaman monokultur sehingga jenis ini tidak dapat hidup di kawasan yang didominasi oleh aktivitas manusia. Suatu yang paling penting bagi kelestarian jenis Kucing Tandang ini adalah mempertahankan kawasan sempadan sungai dan kawasan hutan rawa gambut. Abstrak (Bahasa Malaysia). Kucing Hutan Prionailurus planiceps dianggap sebagai salah satu spesis kucing yang paling terancam di dunia. Taburan spesis ini terhad di selatan Thailand, Semenanjung Malaysia dan di dua kepulauan Sunda yang terbesar iaitu Borneo dan Sumatera. Perkaitannya dengan tanah bencah dan kawasan tanah rendah memberikan tekanan yang tinggi kepada spesis ini kerana habitat ini sangat terancam dengan kemusnahan dan aktiviti penukaran tanah kepada akuakultur. Borneo telah dikenal pasti sebagai kawasan utama untuk Kucing Hutan. Kami mengumpul sejumlah 140 rekod Kucing Hutan di Borneo. Daripada jumlah itu, 50 rekod (Model Seimbang) atau 78 rekod (Model yang ditapis secara spasial) telah digunakan untuk meramal kesesuaian habitat yang berpotensi bagi spesis ini. Walaupun kami meramalkan habitat yang sesuai untuk Kucing Hutan adalah bertaburan di seluruh kawasan tanah rendah Borneo, keputusan yang kami perolehi turut mencadangkan bahawa sesetengah kawasan tanah rendah yang luas adalah tidak sesuai bagi Kucing Hutan. Ini kemungkinan disebabkan kawasan tersebut telah dijadikan ladang kelapa sawit. Antara kawasan penting yang diramal untuk spesis ini adalah hutan di Brunei Darussalam, Taman Negara Sabangau dan kompleks hutan di Kalimantan Tengah, hutan di bahagian utara Kalimantan Timur serta kompleks hutan di bahagian tengah Sabah. Ancaman utama bagi Kucing Hutan adalah penukaran kawasan hutan kepada ladang monokultur yang masih lagi berlaku pada ketika ini. Ini kerana spesis ini dilihat tidak mampu untuk mengguna kawasan yang telah didominasi oleh manusia. Antara langkah penting untuk kelangsungan hidup jangka panjang Kucing Hutan adalah pemuliharaan kawasan hutan yang berhampiran dengan sungai dan hutan paya gambut.

Predictive modelling of Slav settlement location in the middle-Elbe region. – A predictive model has been constructed of the northwestern part of the middle-Elbe region based on an analysis of the relationship between environmental factors and locations of Slav settlement sites. Initially, an evaluation was carried out of the archaeological data on which to base such a model, and the validation area of the model determined. Relevant environmental factors were selected empirically and tested as to their influence on location of settlement site. The determining factors were found to be (a) height above the local drainage line (thalweg) in combination with distance to the nearest surface water, and (b) distance to the nearest ecoboundary. The application of the predictive model with respect to cultural history, as well as its suitability as  a basis for archaeological heritage management are likely to be confirmed by future studies.

Predictive Models werden bereits seit vielen Jahren international eingesetzt und erlangen zunehmend Bedeutung auch
für die praktische Bodendenkmalpflege in Deutschland. Die bisherigen Kenntnisse bezüglich des im Boden verborgenen archäologischen
Bestandes stellen sich, besonders im Rahmen von Großprojekten, fast immer als sehr unvollständig heraus und erschweren damit die
Planung archäologischer Maßnahmen. Archäoprognosekarten liefern Informationen über das archäologische Potenzial eines Raumes und
stellen damit ein wertvolles Planungsinstrument für die Bodendenkmalpflege dar.
Grundlage dieser Kartierungen ist eine umfangreiche archäologische und landschaftsbezogene Datensammlung in einem geographischen
Informationssystem (GIS). Auf Basis verschiedener statistischer Methoden können archäologische Verdachtsflächenkarten erstellt werden.
An Hand von Beispielen aus Nordrhein-Westfalen und dem Land Brandenburg erläutert dieser Beitrag einen dieser leistungsfähigen
statistischen Ansätze. Die Interpretationsmöglichkeiten der Verdachtsflächenkarten für die archäologische Forschung, mögliche
einschränkende Faktoren sowie ihr Einsatz in der Bodendenkmalpflege werden diskutiert.

Abstrak— Kurangnya interaksi, aktivitas dan komunikasi dalam e-learning membuat proses belajar terkendala. Sementara itu menurut data dari alexa (2011) jejaring sosial Facebook, memiliki pertumbuhan jumlah pengguna yang paling signifikan dan tercatat pada peringkat 2 situs yang paling banyak dikunjungi di dunia dan peringkat 1 di Indonesia. Data ini menunjukkan tingginya interaksi, aktivitas dan komunikasi dalam Facebook itu sendiri. Permasalahannya bagaimana meningkatkan dan memanfaatkan jejaring sosial Facebook untuk tujuan pembelajaran. Dalam penelitian ini akan dibuat LMS dengan pendekatan teori Collaborative Learning pada platform Facebook menggunakan Facebook Application Programing Interface (API). Hasil dari penelitian ini menghasilkan LMS yang terintegrasi dengan jejaring sosial Facebook dan berjalan pada canvas Facebook Apps. Kata Kunci— E-learning, Learning Management System, Collaborative Learning, Facebook API, Facebook Apps.

This paper proposes a design of hierarchical fuzzy inference tree (HFIT). An HFIT produces an optimum tree-like structure, i.e., a natural hierarchical structure that accommodates simplicity by combining several low-dimensional fuzzy inference systems (FISs). Such a natural hierarchical structure provides a high degree of approximation accuracy. The construction of HFIT takes place in two phases. Firstly, a nondominated sorting based multiobjective genetic programming (MOGP) is applied to obtain a simple tree structure (a low complexity model) with a high accuracy. Secondly, the differential evolution algorithm is applied to optimize the obtained tree's parameters. In the derived tree, each node acquires a different input's combination, where the evolutionary process governs the input's combination. Hence, HFIT nodes are heterogeneous in nature, which leads to a high diversity among the rules generated by the HFIT. Additionally, the HFIT provides an automatic feature selection because it uses MOGP for the tree's structural optimization that accepts inputs only relevant to the knowledge contained in data. The HFIT was studied in the context of both type-1 and type-2 FISs, and its performance was evaluated through six application problems. Moreover, the proposed multiobjective HFIT was compared both theoretically and empirically with recently proposed FISs methods from the literature, such as McIT2FIS, TSCIT2FNN, SIT2FNN, RIT2FNS-WB, eT2FIS, MRIT2NFS, IT2FNN-SVR, etc. From the obtained results, it was found that the HFIT provided less complex and highly accurate models compared to the models produced by the most of other methods. Hence, the proposed HFIT is an efficient and competitive alternative to the other FISs for function approximation and feature selection.

—The current study investigates the behaviour of time-varying parameters that are based on the score function of the predictive model density at time t. The mechanism to update the parameters over time is the scaled score of the likelihood function. The results revealed that there is high persistence of time-varying, as the location parameter is higher and the skewness parameter implied the departure of scale parameter from the normality with the unconditional parameter as 1.5. The results also revealed that there is a perseverance of the leptokurtic behaviour in stock returns which implies the returns are heavily tailed. Prior to model estimation, the White Neural Network test exposed that the stock price can be modelled by a GAS model. Finally, we proposed further researches specifically to model the existence of time-varying parameters with a more detailed model that encounters the heavy tail distribution of the series and computes the risk measure associated with the returns.

Evaporation, one of the fundamental components of the hydrology cycle, is differently influenced by various meteorological variables in different climatic regions. The accurate prediction of evaporation is essential for multiple water resources engineering applications, particularly in developing countries like Iraq where the meteorological stations are not sustained and operated appropriately for in situ estimations. This is where advanced methodologies such as machine learning (ML) models can make valuable contributions. In this research, evaporation is predicted at two different meteorological stations located in arid and semi-arid regions of Iraq. Four different ML models for the prediction of evaporation-the classification and regression tree (CART), the cascade correlation neu-ral network (CCNNs), gene expression programming (GEP), and the support vector machine (SVM)-were developed and constructed using various input combinations of meteorological variables. The results reveal that the best predictions are achieved by incorporating sunshine hours, wind speed, relative humidity, rainfall, and the minimum, mean, and maximum temperatures. The SVM was found to show the best performance with wind speed, rainfall, and relative humidity as inputs at Station I (R 2 = .92), and with all variables as inputs at Station II (R 2 = .97). All the ML models performed well in predicting evaporation at the investigated locations. ARTICLE HISTORY

Predictive modeling for high probability zones for detection of submerged prehistoric sites remains an integral part of this subdiscipline. It typically relies on a combination of physiographic parameters detectable in offshore zones and known archaeological trends, most of which lie onshore, not offshore. These models are not always temporally and spatially scaled to account for changes in cultural practices and/or baseline ecological conditions associated with climate changes, however. To account better for these potential variations, we used a Geographic Information Systems (GIS) based spatial statistical analysis method. Our study seeks evidence of variation in site correlations to environmental patterns to detect spatial and temporal variation in site locations, and for where site patterns depart from robust quantitative correlations to ecological conditions. Our results support a hypothesis that sites do tend to correlate to certain ecological parameters, but these effects vary across the study area and through time. We also found that some site groups did not correlate well to any particular set of hypothetically favorable ecological conditions, suggesting that other undetected variables played significant roles in those site locations. These may be undetected/untested ecological variables or may represent culturally-based decision making with non-linear relationships with ecological conditions.