‘Leveraging on CSR as a tool of brand communication based on consumer’s perception with extrapolation from a novel 3-factor model

Artificial neural networks (ANN) have become an established discipline and have gained extensive interest within chemical engineering. In recent years, research effort has focused on the use of "hybrid artificial neural networks" (HANN) models that combine both the deterministic and the ANN elements. Several methods have been proposed for combining ANN with first principle relations. In this thesis, a new hybrid scheme, which is similar to that developed by Kasprow for a space-independent and time-dependent fed-batch microbial reactor, was developed for a space-dependent steady-state enzymatic reactor. This scheme combines ANN with mass balances and assumed rate expressions. It was shown that this new hybrid scheme performed significantly better than both black-box ANN model and the hybrid ANN with only mass balance equations. An enzymatic tubular membrane reactor (TMR) was selected as a case study due to the availability of a reliable deterministic/computational model, which can provide simulated process data as needed, as well as its potential industrial importance. Also, two modeling schemes were developed, a fully 'black box' model (BANN), based on ANN technique only, and a simple hybrid model, combining ANN with mass balances (HANN1). Qualitative and quantitative comparisons of the predicted profiles of the above three modeling schemes indicated that the new hybrid scheme (HANN2) performed better than the other two schemes. As a result of adding biochemical knowledge, in the form of mass balances and simplified rate expressions, the new hybrid scheme allowed the process data to be interpolated and extrapolated more accurately.

Proximity problems is a class of important problemswhich involve estimation of distances between geometricobjects. The nearest neighbor search which is a subsetof proximity problems, arises in numerous fields ofapplications, including Pattern Recognition, Statisticalclassification, Computer vision and etc. In this study, anearest neighbor search is presented to move points inthe plane, while query point is static.The proposed method works in the black-box KDSmodel, in which the points location received at regulartime steps while at the same time, an upper bounddmax is known on the maximum displacement of anypoint at each time step. In this paper, a new algorithmis presented for kinetic nearest neighbor search problemin the black-box model, under assumptions on the distribution of the moving point set P. It has been shownhow the kinetic nearest neighbor will be updated at eachtime step in O(k∆k log n) amortized time, where ∆k isthe k-spread of a point set P .Key words: Computational Geometry, Black Box Model, Kinetic, Nearest Neighbor.

The rainfall-runoff process is one of the most complex hydrological phenomena. Estimating runoff in the basin is one of the main conditions for planning and optimal use of rainfall. Using machine learning models in various sciences to investigate phenomena for which statistical information is available is a helpful tool. This study investigates and compares the abilities of HEC-HMS and TOPMODEL as white box models and adaptive neural fuzzy inference system (ANFIS) and gene expression programming (GEP) as black box models in rainfall-runoff simulation using 5-year statistical data. Using the inputs of rainfall and temperature of the previous day and discharge in the steps of the previous 2 days reduced the prediction error of both models. Examining the role of different parameters in improving the accuracy of simulations showed that the temperature as an effective parameter in cold months reduces the amount of prediction error. A comparison of R2, RMSE, and MBE showed that black box models are more effective forecasting tools. Among the black box models, the ANFIS model with R2 = 0.82 has performed better than the GEP model with R2 = 0.76. For white box models, the HEC-HMS and TOPMODEL had R2 equal to 0.3 and 0.25, respectively.

Many computational techniques which could analyze the behavior of electrical arc have been utilized to estimate the interrupting ability of the circuit breakers. Among them, black-box arc model has been known for valuable tool to simulate dynamic arc-circuit interactions during fault current interruption. Previous research works has shown that the black-box model was efficient tool to analyze AC arc characteristics especially for SF 6 and air circuit breakers. But there's no noticeable investigations about the application of black box model to the DC circuit breakers, which is widely used in DC railway systems and DC microgrid. In this work, we tried to verify the validity of the applications of conventional AC black-box model to the DC electric arc. Several conventional arc models containing Cassie and Schwarz model were analyzed considering the tendency of the dc arcing current and voltage waveform using parametric sweep method. Simulation results showed that the weakness of current AC black-box model for DC arc analysis and compared that which model is more suitable for analyzing the DC arc.

Vessels fuel consumption forecast and trim optimisation: A data analytics perspective

In this paper authors investigate the problem of predicting the fuel consumption of a vessel in real scenario based on data measured by the onboard automation systems. The goal is achieved by exploiting three different approaches: White, Black and Gray Box Models. White Box Models (WBM) are based on the knowledge of the physical underling processes. Black Box Models (BBMs) build upon statistical inference procedures based on the historical data collection. Author proposal is a Gray Box Model (GBM) which is able to exploit both mechanistic knowledge of the underlying physical principles and available measurements. Results on real world data shows that the BBM is able to remarkably improve a state-of-the-art WBM, while the GBM is able to encapsulate the a-priory knowledge of the WBM into the BBM so to achieve the same performance of the latter but requiring less historical data.

Machine learning research focuses on the improvement of prediction performance. Progress was made with black-box models that flexibly adapt to the given data. However, due to their increased complexity, black-box models are more difficult to interpret. To address this issue, techniques for interpretable machine learning have been developed, yet there is still a lack of methods to reliably identify interaction effects between predictors under uncertainty. In this work, we present a model-agnostic hypothesis test for the identification of interaction effects in black-box machine learning models. The test statistic is based on the difference between the variance of the estimated prediction function and a version of the estimated prediction function without interaction effects derived via partial dependence functions. The properties of the proposed hypothesis test were explored in simulations of linear and nonlinear models. The proposed hypothesis test can be applied to any black-box prediction model, and the null hypothesis of the test can be flexibly specified according to the research question of interest. Furthermore, the test is computationally fast to apply, as the null distribution does not require the resampling or refitting of black-box prediction models.

With the build-out of large transport networks utilizing optical technologies, more and more capacity is being made available. Innovations in Dense Wave Division Multiplexing (DWDM) and the elimination of optical-electrical-optical conversions have brought on advances in communication speeds as we move into 10 Gigabit Ethernet and above. Of course, there is a need to encrypt data on these optical links as the data traverses public and private network backbones. Unfortunately, as the communications infrastructure becomes increasingly optical, advances in encryption (done electronically) have failed to keep up. This project examines the use of optical logic for implementing encryption in the photonic domain to achieve the requisite encryption rates. This paper documents the innovations and advances of work first detailed in 'Photonic Encryption using All Optical Logic,' [1]. A discussion of underlying concepts can be found in SAND2003-4474. In order to realize photonic encryption designs, technology developed for electrical logic circuits must be translated to the photonic regime. This paper examines S-SEED devices and how discrete logic elements can be interconnected and cascaded to form an optical circuit. Because there is no known software that can model these devices at a circuit level, the functionality of S-SEED devices in an optical circuit was modeled in PSpice. PSpice allows modeling of the macro characteristics of the devices in context of a logic element as opposed to device level computational modeling. By representing light intensity as voltage, 'black box' models are generated that accurately represent the intensity response and logic levels in both technologies. By modeling the behavior at the systems level, one can incorporate systems design tools and a simulation environment to aid in the overall functional design. Each black box model takes certain parameters (reflectance, intensity, input response), and models the optical ripple and time delay characteristics. These 'black box' models are interconnected and cascaded in an encrypting/scrambling algorithm based on a study of candidate encryption algorithms. Demonstration circuits show how these logic elements can be used to form NAND, NOR, and XOR functions. This paper also presents functional analysis of a serial, low gate count demonstration algorithm suitable for scrambling/encryption using S-SEED devices.

For the fast transient study of power transformers, several types of models are available that can be categorized into three groups: white box models, black box models, and gray box models. The identification of their parameters and their usages make the models different. Since white and black box models cannot cover certain cases, such as the failure analysis of a transformer with unknown geometrical data, a new type of model has recently become important for industry and academic researchers. This model, called the gray box model, is intensely analyzed in the current paper, both theoretically and experimentally. The gray box model conceptually lies between the black and white box models. The outcomes of this work have been verified using a suitable test object. The parameter identification of the model is performed by artificial methods. Some useful new ideas, including applying the Weibull Distribution function in inductance estimation and employing an exponential function for the series resistance of windings, are developed in this paper. The number of unknown parameters to be identified in the gray box model is reduced by applying these ideas, which makes the model more simple and feasible.

System identification technology is an important means of aircraft modeling. In order to make up for the shortcomings of traditional system identification technology, which are complicated in calculation and limited in applicability, this paper studies the aircraft intelligent system identification technology based on black and gray box modeling. The black box model which is based on the recurrent neural network establish the dynamic relationship between the input and output variables of the aircraft; while gray box model treats the motion equation and aerodynamic coefficients of the aircraft separately, assuming that the motion equation is known, and uses an extreme learning machine to establish the forward neural network model of the aerodynamic coefficients. Thus two intelligent models of the aircraft are established, the grey box model is mainly used for model prediction for a long period, and can be utilized for flight simulation, the black box model is mainly used for online model prediction, and can be applied for model predictive control.

With the build-out of large transport networks utilizing optical technologies, more and more capacity is being made available. Innovations in Dense Wave Division Multiplexing (DWDM) and the elimination of optical-electrical-optical conversions have brought on advances in communication speeds as we move into 10 Gigabit Ethernet and above. Of course, there is a need to encrypt data on these optical links as the data traverses public and private network backbones. Unfortunately, as the communications infrastructure becomes increasingly optical, advances in encryption (done electronically) have failed to keep up. This project examines the use of optical logic for implementing encryption in the photonic domain to achieve the requisite encryption rates. In order to realize photonic encryption designs, technology developed for electrical logic circuits must be translated to the photonic regime. This paper examines two classes of all optical logic (SEED, gain competition) and how each discrete logic element can be interconnected and cascaded to form an optical circuit. Because there is no known software that can model these devices at a circuit level, the functionality of the SEED and gain competition devices in an optical circuit were modeled in PSpice. PSpice allows modeling of the macro characteristics of the devices in context of a logic element as opposed to device level computational modeling. By representing light intensity as voltage, 'black box' models are generated that accurately represent the intensity response and logic levels in both technologies. By modeling the behavior at the systems level, one can incorporate systems design tools and a simulation environment to aid in the overall functional design. Each black box model of the SEED or gain competition device takes certain parameters (reflectance, intensity, input response), and models the optical ripple and time delay characteristics. These 'black box' models are interconnected and cascaded in an encrypting/scrambling algorithm based on a study of candidate encryption algorithms. We found that a low gate count, cascadable encryption algorithm is most feasible given device and processing constraints. The modeling and simulation of optical designs using these components is proceeding in parallel with efforts to perfect the physical devices and their interconnect. We have applied these techniques to the development of a 'toy' algorithm that may pave the way for more robust optical algorithms. These design/modeling/simulation techniques are now ready to be applied to larger optical designs in advance of our ability to implement such systems in hardware.

People are becoming more aware of environmental difficulties in this time of globalization. Waste is one such problem, as waste is a major source of pollution in the air, soil and water. It is hoped that all individuals and businesses in Indonesia, including consumers, will increase awareness and concern for the environment as an effort to preserve and restore the earth's ecosystem. Emina products are one example of a company that prioritizes sustainability in its systems and products at PT Technology and Innovation. Emina is considered a product that contributes to a positive reputation and values social responsibility. Green marketing includes all efforts and endeavors directed at promoting, packaging, distributing, consuming, and recalling products through environmental awareness or response. Corporate social responsibility is a condition in which businesses must continue to pay attention to the environment in an effort to reduce environmental problems that have occurred or are occurring. To achieve sustainable growth and performance, corporate social responsibility is an initiative that is integrated into the investment strategy and is considered a corporate social obligation. Brand image, alongside environmental concerns and corporate social responsibility will emerge as an important consideration for companies in the coming years. Consequently, brand image is consumers' and customers' perception of a brand, which in turn inspires trust in the brand. This research highlights the importance of green marketing and environmental accountability in general by building a good perception of the organization's brand within the local community. The purpose of this study is to determine the correlation between brand image (Z) and the impact of corporate social responsibility (Y) and green marketing (X). Uncertainty surrounds most Emina product users who make up the population of this study. In the absence of population information, the sample size for a given study will be calculated based on criteria for an unknown population, resulting in a total of 100 respondents. This research will use a purposive sampling method, which involves selecting a sample from the population under study based on certain criteria. Regression analysis and path analysis will be used to analyze the data. Hypothesis testing, validity and reliability tests, normality and multicollinearity tests, path analysis model fit tests, and SPSS and AMOS Softwer analysis tools will be used for data analysis. According to the findings of this study, green marketing influences consumer perceptions of Emina products favorably, thereby contributing to improved environmental stewardship. Emina is perceived as an ethical and socially responsible entity, which positively impacts its reputation

We study the power of quantum proofs, or more precisely, the power of quantum Merlin-Arthur (QMA) protocols, in two well studied models of quantum computation: the black box model and the communication complexity model. Our main results are obtained for the communication complexity model. For this model, we identify a complete promise problem for QMA protocols, the linear sub-spaces distance problem. The problem is of geometrical nature: each player gets a linear subspace of R/sup m/ and considers the sphere of unit vectors in that subspace. Their goal is to output 1 if the distance between the two spheres is very small (say, smaller than 0.1 /spl middot/ /spl radic/2) and 0 if the distance is very large (say, larger than 0.9 /spl middot/ /spl radic/2). We show that: 1. The QMA communication complexity of the problem is O(logm). 2. The (classical) MA communication complexity of the problem is /spl Omega/(m/sup /spl epsi//) (for some /spl epsi/ > 0). 3. The (standard) quantum communication complexity of the problem is /spl Omega/(/spl radic/m). In particular, this gives an exponential separation between QMA communication complexity and MA communication complexity. For the black box model we give several observations. First, we observe that the block sensitivity method, as well as the polynomial method for proving lower bounds for the number of queries, can both be extended to QMA protocols. We use these methods to obtain lower bounds for the QMA black box complexity of functions. In particular, we obtain a tight lower bound of /spl Omega/(N) for the QMA black box complexity of a random function, and a tight lower bound of /spl Omega/(/spl radic/N) for the QMA black box query complexity of NOR(X/sub 1/,..., X/sub n/). In particular, this shows that any attempt to give short quantum proofs for the class of languages Co - NP have to go beyond black box arguments. We also observe that for any Boolean function G(X/sub 1/,..., X/sub n/), if for both G and 7minus;G there are QMA black box protocols that make at most T queries to the black box, then there is a classical deterministic black box protocol for G that makes 0(T/sup 6/) queries to the black box. In particular, this shows that in the black box model QMA /spl cap/ Co - QMA = P. On the positive side, we observe that any (total or partial) Boolean function G(X/sub 1/,..., X/sub n/) has a QMA black box protocol with proofs of length N that makes only 0(/spl radic/N) queries to the black box. Finally, we observe a very simple proof for the exponential separation (for promise problems) between QMA black box complexity and (classical) MA black box complexity (first obtained by Watrous).

Air quality prediction with black-box (BB) modelling is gaining widespread interest in research and industry. This type of data-driven models work generally better in terms of accuracy but are limited to capture physical, chemical and meteorological processes and therefore accountability for interpretation. In this paper, we evaluated different white-box (WB) and BB methods that estimate atmospheric black carbon (BC) concentration by a suite of observations from the same measurement site. This study involves data in the period of 1st January 2017–31st December 2018 from two measurement sites, from a street canyon site in Mäkelänkatu and from an urban background site in Kumpula, in Helsinki, Finland. At the street canyon site, WB models performed (R2 = 0.81–0.87) in a similar way as the BB models did (R2 = 0.86–0.87). The overall performance of the BC concentration estimation methods at the urban background site was much worse probably because of a combination of smaller dynamic variability in the BC values and longer data gaps. However, the difference in WB (R2 = 0.44–0.60) and BB models (R2 = 0.41–0.64) was not significant. Furthermore, the WB models are closer to physics-based models, and it is easier to spot the relative importance of the predictor variable and determine if the model output makes sense. This feature outweighs slightly higher performance of some individual BB models, and inherently the WB models are a better choice due to their transparency in the model architecture. Among all the WB models, IAP and LASSO are recommended due to its flexibility and its efficiency, respectively. Our findings also ascertain the importance of temporal properties in statistical modelling. In the future, the developed BC estimation model could serve as a virtual sensor and complement the current air quality monitoring. Main findingsWhite-box models are preferred over black-box models in estimating black carbon because they are closer to physics-based models, and it is easier to spot the relative importance of the predictor variable. The black carbon model could serve as a virtual sensor integrating into air quality network in support with real measurements, so as to complement the current air quality index.

In recent years, deep neural networks (DNN) have become a highly active area of research, and shown remarkable achievements on a variety of computer vision tasks. DNNs, however, are known to often make overconfident yet incorrect predictions on out-of-distribution samples, which can be a major obstacle to real-world deployments because the training dataset is always limited compared to diverse real-world samples. Thus, it is fundamental to provide guarantees of robustness to the distribution shift between training and test time when we construct DNN models in practice. Moreover, in many cases, the deep learning models are deployed as black boxes and the performance has been already optimized for a training dataset, thus changing the black box itself can lead to performance degradation. We here study the robustness to the geometric transformations in a specific condition where the black-box image classifier is given. We propose an additional learner, REinforcement Spatial Transform learner (REST), that transforms the warped input data into samples regarded as in-distribution by the black-box models. Our work aims to improve the robustness by adding a REST module in front of any black boxes and training only the REST module without retraining the original black box model in an end-to-end manner, i.e. we try to convert the real-world data into training distribution which the performance of the black-box model is best suited for. We use a confidence score that is obtained from the black-box model to determine whether the transformed input is drawn from in-distribution. We empirically show that our method has an advantage in generalization to geometric transformations and sample efficiency.