Deep learning in exchange markets
یادگیری عمیق در بازارهای ارز-2019
We present the implementation of a short-term forecasting system of price movements in exchange mar- kets using market depth data and a systematic procedure to enable a fully automated trading system. Three types of Deep Learning (DL) Neural Network (NN) methodologies are trained and tested: Deep NN Classifier (DNNC), Long Short-Term Memory (LSTM) and Convolutional NN (CNN). Although the LSTM is more suitable for multivariate time series analysis from a theoretical point of view, test results indicate that the CNN has on average the best predictive power in the case study under analysis, which is the UK to Win Horse Racing market during pre-live stage in the world’s most relevant betting exchange. Implica- tions from the generalized use of automated trading systems in betting exchange markets are discussed.
Keywords: Deep learning | Betting exchange | Market depth | Classification
Machine Learning based Digital Twin Framework for Production Optimization in Petrochemical Industry
چارچوب دوقلوی دیجیتال مبتنی بر یادگیری ماشین برای بهینه سازی تولید در صنعت پتروشیمی-2019
Digital twins, along with the internet of things (IoT), data mining, and machine learning technologies, offer great potential in the transformation of today’s manufacturing paradigm toward intelligent manufacturing. Production control in petrochemical industry involves complex circumstances and a high demand for timeliness; therefore, agile and smart controls are important components of intelligent manufacturing in the petrochemical industry. This paper proposes a framework and approaches for constructing a digital twin based on the petrochemical industrial IoT, machine learning and a practice loop for information exchange between the physical factory and a virtual digital twin model to realize production control optimization. Unlike traditional production control approaches, this novel approach integrates machine learning and real-time industrial big data to train and optimize digital twin models. It can support petrochemical and other process manufacturing industries to dynamically adapt to the changing environment, respond in a timely manner to changes in the market due to production optimization, and improve economic benefits. Accounting for environmental characteristics, this paper provides concrete solutions for machine learning difficulties in the petrochemical industry, e.g., high data dimensions, time lags and alignment between time series data, and high demand for immediacy. The approaches were evaluated by applying them in the production unit of a petrochemical factory, and a model was trained via industrial IoT data and used to realize intelligent production control based on real-time data. A case study shows the effectiveness of this approach in the petrochemical industry.
Keywords: digital twin | machine learning | internet of things | petrochemical industry | production control optimization
Land use/land cover dynamics using landsat data in a gold mining basin-the Ankobra, Ghana
پویایی استفاده از زمین / پوشش زمین با استفاده از داده های زمین در یک حوضه استخراج طلا - آنکارا ، غنا-2019
The Ankobra River basin, which forms part of the Southern-Western River System of Ghana, has become a hub for both large and small-scale mining activities. Some of the activities of these mines threatens the sustainability of its resources in the basin. A study in the basin which focuses on changes in LULC patterns in the basin over the past years is therefore paramount to understand the changes that has taken place and its potential impact on resources in the basin. This study assessed the pattern of Land use/cover changes, and the possible drivers of change, in the basin. The study used multi-spectral Landsat images of 30 m resolution for the years 1991, 2002, 2008 and 2016. The Semi-Automatic Classification Plugin (SCP) in QGIS was used for atmospheric correction, image classification using the spectral angle mapping algorithm, and post processing. The overall difference between the reference and the classified map of 2016 was 5.5% with the overall quantity, exchange and shift components as 1.5%, 3% and 1% respectively. Findings from the study show that, the Ankobra River basin has witnessed noticeable changes over the 25-year study period. Closed forest which occupied 40.4% of the total basin area in 1991 reduced drastically to 22.8% in 2016. The dominant LULC change patterns in the basin are from Closed Forest to Open Forest; Open Forest to Farmland, Settlement/Bare land and Mining Area; and the increase in surface area of Water consequently resulting from the increase in Mining Area. Mining activities, particularly illegal mining was identified to be the dominant driver of deforestation in the Ankobra Basin between the years 2008 and 2016 where mining activities in the basin sharply increased.
Keywords: Land use/land cover (LULC) | Remote sensing | Ankobra River Basin | Galamsey | Supervised classification | Deforestation
A local and global event sentiment based efficient stock exchange forecasting using deep learning
پیش بینی بورس اوراق بهادار کارآمد مبتنی بر احساسات محلی و جهانی با استفاده از یادگیری عمیق-2019
Stock exchange forecasting is an important aspect of business investment plans. The customers prefer to invest in stocks rather than traditional investments due to high profitability. The high profit is often linked with high risk due to the nonlinear nature of data and complex economic rules. The stock markets are often volatile and change abruptly due to the economic conditions, political situation and major events for the country. Therefore, to investigate the effect of some major events more specifically global and local events for different top stock companies (country-wise) remains an open research area. In this study, we consider four countries- US, Hong Kong, Turkey, and Pakistan from developed, emerging and underdeveloped economies’ list. We have explored the effect of different major events occurred during 2012–2016 on stock markets. We use the Twitter dataset to calculate the sentiment analysis for each of these events. The dataset consists of 11.42 million tweets that were used to determine the event sentiment. We have used linear regression, support vector regression and deep learning for stock exchange forecasting. The performance of the system is evaluated using the Root Mean Square Error (RMSE) and Mean Absolute Error (MAE). The results show that performance improves by using the sentiment for these events.
Keywords: Stock prediction | Regression | Deep learning | Event sentiment
SANUB: یک روش جدید برای به اشتراک گذاری و تحلیل اخبار با استفاده از بلاکچین (زنجیره بلوکی)
سال انتشار: 2019 - تعداد صفحات فایل pdf انگلیسی: 5 - تعداد صفحات فایل doc فارسی: 14
میلیون ها خبر به صورت روزانه بین مردم رد و بدل میشود. با ظهور اینترنت، روش اخبار گسترده تغییر کرده و سریعتر شده است، با این حال مشکلات زیادی را ایجاد کرده است. برای مثال، افزایش سرعت انتشار اخبار منجر به افزایش سرعت انتشار اخبار جعلی میشود. اخبار تقلبی میتواند تاثیر زیادی بر جوامع داشته باشد علاوه بر این، وجود یک نهاد مرکزی، مانند آژانسهای خبری، میتواند منجر به تقلب در فرآیند پخش خبر شود، به عنوان مثال تولید اخبار تقلبی و انتشار. از آنجا که تکنولوژی بلاکچین یک شبکه غیر متمرکز قابلاطمینان فراهم میکند، میتواند برای انتشار اخبار استفاده شود. علاوه بر این، بلاکچین با کمک برنامههای کاربردی غیر متمرکز و قراردادهای هوشمند میتواند سکویی را فراهم کند که در آن اخبار تقلبی را می توان از طریق مشارکت عمومی تشخیص داد. در این مقاله، ما روش جدیدی را برای به اشتراک گذاری و تحلیل اخبار به منظور شناسایی اخبار جعلی با استفاده از بلاکچین به نام SANUB پیشنهاد کردیم. SANUB ویژگیهایی مانند انتشار ناشناس اخبار، ارزیابی اخبار، ارزیابی گزارشگر، تشخیص اخبار تقلبی و اثبات مالکیت اخبار را ارایه میدهد. نتایج تحلیل ما نشان میدهد که SANUB بهتر از روشهای موجود عمل کرده است.
|مقاله ترجمه شده|
Wavelet-based clod segmentation on digital elevation models of a soil surface with or without furrows
تقسیم بندی کلود مبتنی بر موجک در مدلهای ارتفاع دیجیتال سطح خاک با یا بدون برس-2019
Soil surface roughness is a key factor for our understanding and modelling of geomorphologic processes related to exchanges of soil, water and gas. It has an impact on soil properties and tillage outcome. Soil surface roughness can be characterized both globally and locally. The interest of clod segmentation is to allow for both characterizations. Segmenting clods on a digital elevation model (DEM) of a soil surface is a complex problem because soil surfaces are complex surfaces of several level of roughness and because considering elevations results in smooth and poorly contrasted images. However, a DEM in 3 dimensions gathers more information than a profile of 1 dimension or an image of 2 dimensions. Multiresolution analysis has shown interest for roughness analysis of complex surfaces. We have used it to introduce a new approach for soil roughness analysis and to lay the foundations for clod segmentation. In this paper, we propose a complete wavelet-based approach for accurate clod contour delineation. It relies on several steps: detecting clods on the surface approximations by a supervised detection of local maxima, validating and merging the detections by shape and overlap tests, delineating the clod contours by intersecting locally the soil surface elevations with the estimated plane of the clod base and validating the contours by detecting and correcting the wrong patterns, with statistical pattern recognition. This segmentation method was evaluated in several roughness conditions, made in the laboratory, by comparison with other segmentation method. An indicator of goodness of agreement was introduced for this purpose. This wavelet-based segmentation method showed robustness to the presence of furrows and to the smoothing by rainfall and showed ability to retrieve clod diameters.
Keywords: Random roughness | 3D digital elevation model | Multiresolution analysis | Statistical pattern recognition | Clod size
Entrepreneurship in marketing: Socializing partners for brand governance in EM firms
کارآفرینی در بازاریابی: اجتماعی کردن شرکای حاکمیت برند در شرکتهای EM-2019
We present case studies of opportunity exploration and exploitation in two small, established Entrepreneurial Marketing firms who actively build online and offline partnerships with a range of stakeholders. We use effectuation principles to explore the use of the relational norms of trust, selection, solidarity, information exchange and flexibility within the firms, considering these as mechanisms used by the organization to socialise their partners. We examine the way firms vary their emphasis on trust, selection, solidarity and information exchange in line with the nature and importance of the partnership, and how each contributes to opportunity exploration. We also examine how flexibility enables the exploitation of opportunities validated as having potential. Finally, we present a theoretical framework based on our results along with propositions for future research. Contributions are made to entrepreneurial marketing, effectuation, and brand governance literatures.
Keywords: Entrepreneurial marketing | Socialization | Effectuation | Relational norms | Brand governance | SMEs
Cooperative enhanced scatter search with opposition-based learning schemes for parameter estimation in high dimensional kinetic models of biological systems
جستجوی پراکنده افزایش یافته مشارکتی با طرح یادگیری مبتنی بر مخالفت برای تخمین پارامتر در مدل های جنبشی ابعادی بالا از سیستم های بیولوژیکی-2019
Industrial bioprocesses development nowadays is concerned with producing chemicals using yeast, bac- teria and therapeutic proteins in mammalian cells. This involves the utilization of microorganism cells as factories and re-engineering them in silico . The tools that could facilitate this process are known as the ki- netic models. Kinetic models of cellular metabolism are important in assisting researchers to understand the rational design of biological systems, predicting metabolites production, and improving bio-products development. However, the most challenging task in model development is parameter estimation, which is the process of identifying an unknown value of model parameters which provides the best fit between the model output and a set of experimental data. Due to the increased complexity and high dimension- ality of the models, which are extremely nonlinear and contain large numbers of kinetic parameters, parameter estimation is known to be difficult and time-consuming. This study proposes a cooperative enhanced scatter search with opposition-based learning schemes (CeSSOL) for parameter estimation in large-scale biology models. The method was executed in parallel with the proposed cooperative mecha- nism in order to exchange information (kinetic parameters) between individual threads. Each thread con- sists of different parameters settings that enhance the systemic properties in obtaining the global min- imum. The performance of the proposed method was assessed against two large-scale microorganisms models using mammalian and bacteria cells. The results revealed that the proposed method recorded faster computation time compared to other methods. The study has also demonstrated that the proposed method can be used to provide more accurate and faster estimation of kinetic models, indicating the potential benefits of utilizing this method for expert systems of industrial biotechnology.
Keywords: Parameter estimation | Metabolic engineering | Kinetic model | Opposition-based learning | Global optimization | Cooperative metaheuristic
Blockchain for cloud exchange: A survey
بلاکچین برای تبادل ابر: یک بررسی-2019
Compared with single cloud service providers, cloud exchange provides users with lower price and flexible options. However, conventional cloud exchange markets are suffer- ing from a number of challenges such as central architecture being vulnerable to mali- cious attacks and cheating behaviours of third-party auctioneers. The recent advances in blockchain technologies bring the opportunities to overcome the limitations of cloud ex- change. However, the integration of blockchain with cloud exchange is still in infancy and extensive research efforts are needed to tackle a number of research challenges. To bridge this gap, this paper presents an overview on using blockchain for cloud exchange. In par- ticular, we first give an overview on cloud exchange. We then briefly survey blockchain technology and discuss the issues on using blockchain for cloud exchange in aspects of security, privacy, reputation systems and transaction management. Finally, we present the open research issues in this promising area.
Keywords: Industrial internet of things | Cloud exchange | Blockchain | Decentralization | Reputation systems
Multi-output bus travel time prediction with convolutional LSTM neural network
پیش بینی زمان سفر با اتوبوس چند خروجی با شبکه عصبی LSTM حلقوی-2019
Accurate and reliable travel time predictions in public transport networks are essential for delivering an attractive service that is able to compete with other modes of transport in urban areas. The traditional application of this information, where arrival and departure predictions are displayed on digital boards, is highly visible in the city landscape of most modern metropolises. More recently, the same information has become critical as input for smart-phone trip planners in order to alert passengers about unreachable connections, alternative route choices and prolonged travel times. More sophisticated Intelligent Transport Systems (ITS) include the predictions of connection assurance, i.e. an expert system that will decide to hold services to enable passenger exchange, in case one of the services is delayed up to a certain level. In order to operate such systems, and to ensure the confidence of passengers in the systems, the infor- mation provided must be accurate and reliable. Traditional methods have trouble with this as congestion, and thus travel time variability, increases in cities, consequently making travel time predictions in urban areas a non-trivial task. This paper presents a system for bus travel time prediction that leverages the non-static spatio-temporal correlations present in urban bus networks, allowing the discovery of com- plex patterns not captured by traditional methods. The underlying model is a multi-output, multi-time- step, deep neural network that uses a combination of convolutional and long short-term memory (LSTM) layers. The method is empirically evaluated and compared to other popular approaches for link travel time prediction and currently available services, including the currently deployed model at Movia, the regional public transport authority in Greater Copenhagen. We find that the proposed model significantly outper- forms all the other methods we compare with, and is able to detect small irregular peaks in bus travel times very quickly.
Keywords: Bus travel time prediction | Intelligent Transport Systems | Convolutional neural network (CNN) | Long short-term memory (LSTM) | Deep learning