Search

The novelty of this research lies in the integration of machine learning and artificial intelligence techniques into stochastic DEA models. While traditional DEA models have been widely used to measure the efficiency of decision-making units, they may not be able to capture complex and nonlinear relationships between inputs and outputs. By integrating advanced machine learning and AI techniques, this research aims to improve the accuracy, stability, and interpretability of stochastic DEA models, providing decision-makers with more reliable and actionable insights. Moreover, this research explores several novel approaches, including the integration of deep learning techniques, ensemble learning, dynamic stochastic DEA models, and explainable AI, to improve the performance of stochastic DEA models. These approaches have the potential to enhance the accuracy of efficiency scores, increase the stability of the model, provide more actionable insights, and improve the model's interpretability. By integrating these approaches into stochastic DEA models, this research aims to provide a comprehensive and effective solution to the problem of measuring the efficiency of decision-making units. This approach has not been explored extensively in the literature, and thus represents a novel and innovative approach to addressing this important research problem.

The novelty of integrating neural networks (NNs) into the stochastic DEA model lies in the ability to address some of the limitations of traditional DEA models and improve the accuracy and efficiency of efficiency measurement and prediction. By integrating NNs, the stochastic DEA model can capture the complex and non-linear relationships between the input and output variables of the decision-making units (DMUs) and handle uncertainty in the input and output data. This is achieved by using the NN to estimate the distribution of the input and output data and then using the stochastic DEA model to calculate the efficiency scores based on these estimated distributions. Furthermore, the integration of NNs into the stochastic DEA model allows for the development of hybrid models that combine the strengths of both techniques. For example, some researchers have proposed using genetic algorithms or other optimization techniques to optimize the input and output weights of the stochastic DEA model, which are then used to calculate the efficiency scores based on the estimated distributions from the NN. This results in a more accurate and efficient efficiency measurement and prediction model. Another novelty of integrating NNs into the stochastic DEA model is the potential for enhancing the interpretability of the model. While NNs are often considered as black-box models, several methods have been proposed to enhance the interpretability of NN-based stochastic DEA models. These methods include using feature importance analysis or visualization techniques to identify the most important input and output variables that contribute to the efficiency scores. Overall, the integration of NNs into the stochastic DEA model represents a novel approach to addressing the limitations of traditional DEA models and improving the accuracy and efficiency of efficiency measurement and prediction under uncertainty. The development of hybrid models and methods to enhance interpretability further add to the novelty and potential impact of this research.

The research on efficient optimization algorithms for machine learning is novel because it addresses several gaps in previous research and proposes new solutions to improve the efficiency and accuracy of machine learning models. Firstly, the proposed research focuses on developing more efficient algorithms for large-scale deep learning. While there have been many optimization algorithms proposed for deep learning, the proposed research aims to develop new algorithms that can handle the complexity and scale of these models and improve their efficiency. Secondly, the proposed research aims to explore the effectiveness of optimization algorithms for different types of machine learning tasks. While many studies have focused on deep learning, the proposed research aims to evaluate the effectiveness of optimization algorithms for other types of machine learning tasks, such as reinforcement learning, unsupervised learning, and semi-supervised learning. Thirdly, the proposed research aims to develop optimization algorithms that can handle noisy and incomplete data, which is a significant challenge for machine learning models. The proposed research aims to develop algorithms that can handle noisy and incomplete data and improve the accuracy of machine learning models. Fourthly, the proposed research aims to develop optimization algorithms that can handle non-convex objective functions. While some optimization techniques have been proposed for non-convex optimization, the proposed research aims to develop new algorithms that can handle these functions and improve the accuracy of machine learning models. The proposed research aims to investigate the trade-off between optimization efficiency and model performance. While previous research has explored this trade-off to some extent, the proposed research aims to develop algorithms that can balance these factors and optimize both efficiency and performance. The proposed research is novel because it addresses several gaps in previous research and proposes new solutions to improve the efficiency and accuracy of machine learning models for various tasks, including classification, regression, and clustering. By developing new algorithms and evaluating their effectiveness for different types of machine learning tasks, the proposed research can advance the field of machine learning and improve the accuracy and efficiency of machine learning models.

Efficiency assessment is vital to assessing decision-making units (DMUs) in numerous sectors. DEA is a prominent non-parametric efficiency assessment tool. Traditional DEA models assume deterministic inputs and outputs, ignoring real-world uncertainties and variability. To improve efficiency estimation, stochastic and robust optimization approaches can be integrated into DEA models. To improve efficiency estimation, we present a stochastic and robust optimization framework incorporating DEA. Probabilistic inputs and outputs allow stochastic optimization to account for uncertainty. The model can capture data variability and create stochastic DMU efficiency scores by adding probability distributions. For data uncertainties and outliers, the DEA model uses robust optimization. Robust optimization considers worst-case scenarios and minimizes extreme observations on efficiency estimation. This makes efficiency scores more resilient to data outliers and noise. DEA models benefit from stochastic and resilient optimization. First, considering data uncertainties and fluctuations improves DMU efficiency representation. Second, eliminating outliers and extreme observations improves efficiency estimation. Third, efficiency scores help decision-makers make better, more informed choices. A case study in a specific industry shows the framework's efficacy. We compare classic and integrated stochastic-robust DEA model outcomes. The integrated model provides more accurate and dependable efficiency estimates, helping decision-makers understand DMU performance. DEA models with stochastic and resilient optimization increase efficiency estimation. By considering uncertainties and outliers, this paradigm helps decision-makers evaluate DMUs in many sectors more accurately and reliably.

Decision making based on data driven deals with a large amount of data will evaluate the process's effectiveness. Evaluate effectiveness in this paper is measure of performance efficiency of data envelopment analysis (DEA) method in this study is the approach with uncertainty problems. This study proposed a new method called the robust stochastic DEA (RSDEA) to approach performance efficiency in tackling uncertainty problems (i.e., stochastic and robust optimization). The RSDEA method develops to combine the stochastics DEA (SDEA) formulation method and Robust Optimization. The numerical example demonstrates the performance efficiency of the proposed formulation method, with the results performing confirmed that the efficiency value is 89%.

The most important element of higher education was students, therefore every university must continue to improve services in the future, and one of them was by using decision support. This case could be done by utilizing the University of Big Data. Predicting the number of prospective students in higher education was done by utilizing data mining and multiple linear regression approaches. By using 2 independent variables, namely administration costs (X1), accreditation score (X2), and the number of students who was registered each year as dependent variable (Y). For the test data, it used database for the last 13 years. By using multiple linear regression, the intercept value was sought and the coefficient of determination until the regression coefficient was obtained with the equation Y = 45.28 + -0.02.X1 + 121.58.X2, noted that if X2 was constant, the increasing of one unit was in X1 would have the effect of increasing -0.02 units on Y. Secondly, if X1 was constant, the increasing of one unit was in X2, would have the effect of increasing 121.58 units in Y. Thirdly, if X1 and X2 were equal to zero, the magnitude of Y was 45.28 units. Therefore, the proposed approach could be provided the acceptable predictive results.

Diseases and pests on cardamom plants is one of the diseases and pests that can seriously attack cardamom plants. Cardamom plant diseases and pests can be diagnosed through the symptoms that are currently being experienced by the cardamom or through its clinical picture, through these symptoms an expert system can be made to make a diagnosis. An expert system is a system that seeks to adopt human knowledge to a computer that is built to solve problems like an expert. The expert system made in carrying out the diagnosis uses the Naïve Bayes method. This method is a simple probabilistic-based prediction technique based on the application of Bayes' rules with the assumption of strong (naive) independence. In other words, in Naïve Bayes the model used is an “independent feature model”. This expert system was built using PHP and MySQL programming as a database. In this expert system the types of cardamom disease diagnosed consisted of aphids, leaf-eating caterpillars, stem borers, fruit and roots, leaf beetles, mosaics, late blight, root rot, and fungus, which consisted of 33 symptoms. While the results of the diagnosis will inform about the results of the diagnosis containing a list of symptoms entered, information on the results of regulations regarding diseases and pests that are attacking cardamom plants and information about possible treatments that can be carried out and treatment solutions.

Tujuan dari penelitian ini adalah untuk mengetahui identifikasi kandungan zat pengawet berbahaya boraks dan formalin pada makanan. Metode yang digunakan untuk mengidentifikasi kandungan zat berbahaya pada makanan dengan menggunakan Certainty Factor dengan teknik pemberian bobot pada setiap premis (gejala) hingga memperoleh persentase keyakinan untuk mengidentifikasi makanan yang mengandung formalin dan boraks. Hasil penelitian ini adalah Kandungan boraks pada makanan, dari 4 sampel makanan (100%) yaitu 4 sampel atau seluruh sampel tidak mengandung boraks dengan persentase sebesar 100%. Kandungan formalin pada makanan, dari 4 sampel makanan (100%) yaitu ada 2 sampel makanan positif mengandung formalin dengan persentase sebesar 50% dan ada 2 makanan negative mengandung formalin dengan persentase sebesar 50%. Dari hasil pemeriksaan menggunakan spektrofoto meter UV-VIS kadar formalin yang terendah terdapat pada sampel (Ikan Segar) dengan nilai 0,6631 mg/l. Kadar formalin yang tertinggi terdapat pada sampel C (Mi Bakso) dengan nilai 1,7140 mg/l.

J. City Residence provides subsidized housing loans facilities for people who earn below the average. The number of credit applicants with different criteria requires carefulness of the Credit Analyst in making decisions. This problem can be solved by building a Decision Support System (DSS) in determining the provision of subsidized mortgage loans using the Simple Additive Weighting (SAW) method. The criteria used are house condition (cost attribute), income (cost attribute), employment (benefit attribute), credit history (benefit attribute) and marital status (benefit attribute). The process is to normalize the credit applicant value matrix, then multiply the results of the normalization by the weight value. If the result of the calculation is above the credit line is not feasible, then the applicant is declared eligible to receive credit. Application can be used to help to determine the eligibility of consumers in obtaining subsidized housing loans with the SAW method in J. City Residence by Capital Property Housing.

Online information needs have evolved in the real direction. These needs include the latest information, government services, and commercial products. The research question is how to describe and optimize keyword research with the allintitle technique on the google search engine. The development method used in this research is the prototype method because it is considered able to be evaluated directly on the user. The system testing is done for 3 months by placing keywords on several websites on Google. The conclusion that can be taken is to use the allintitle technique, the search results for the web are easier to find. And this web-based allintitle technique can overcome the challenges of captcha verification from the Google search engine. Keywords: Allintitle, Google's Search Engine, Keyword competition.

Herpes Zoster disease is one of the most difficult skin diseases to treat and everyone can experience it, the characteristic of this disease is marked by unilateral vesicular grouped with pain marked with radicular around the dermatome. This study aims to build bayes theorem method using Visual Basic 2008 as a tool to diagnose skin diseases. With this application does not have to wait long to anticipate treatment quickly and precisely. How to use this application is the admin input questions in the form of symptoms experienced by the patient, then the system will process all the patient answers using Bayes method a nd the system will output the results of the diagnosis of the type of disease. The built system can help the patient in knowing the type of illness that is being suffered by the patient and in accordance with the expert analysis of skin diseases.

This research makes the system in decision making for tourists who will visit an area. In making decision decision systems using the Bayes algorithm. Bayes’s theorem was put forward by a 1763 British Presbyterian minister named Thomas Bayes. In this study Bayes’s theorem is used to calculate the probability of occurrence of an event based on the influence obtained from the observations. In this paper, the Naive Bayes algorithm is used to support the Decision Making System (DSS) to determine the best and strategic location in determining hotel accommodations, especially for tourists who are associated with having mediocre funds in traveling to Lombok Island. In this study applied a number of criteria and data sets obtained from questionnaires from some visiting tourists. There are 3 variables including Hotel Prices, Distance, and the presence or absence of transportation from tourist destinations, from the three variables, to the training dataset. So that in the future there will be questions in the form of data testing to answer the decision whether the hotel can be take or not by looking at the highest probability value.

Security of data text file on a computer can be done by utilizing encryption and decryption techniques. One technique is encryption and decryption of data encryption system text file with cryptography. Cryptography is the science or art to randomize and secure the message to avoid the manipulation of data by performing encryption and decryption on the data of the text file. Cryptographic algorithm consists of several types of one example is the algorithm Rivest Shamir Adleman (RSA). The RSA algorithm is one of asymmetric cryptography technique ie riptografi modern technique that has two keys of the public key used for encryption and a private key techniques to make the process deskripsi. The purpose of this research is to design and implement RSA cryptography algorithm on text file data security. The results of this design is a dekstop based software. This research is expected to provide benefits to the users, individually as well for other organizations.

This research examines how to design an application of an expert system to diagnose Syzygium aqueum (burm.f.alston) plant diseases so that it is easier to detect diseases in the Syzygium aqueum (burm.f.alston) plants. This study uses the Bayes method to be more precise because it is based on the probability value of disease symptoms that arise in Syzygium aqueum (burm.f.alston) plants. In this research, the authors used a web-based application with the Bayes method. The results or outputs from this application provide the probability value of disease certainty and then the hypothesis is chosen with the largest value.

PC computer is a flexible electronic device and can receive and process data quickly and accurately, and can also store data for a long time. There is a time when the PC hardware is damaged. So that it annoys the user. In connection with the problems that occur, the researchers aim to build an Expert System application to diagnose damage to PC computers using the Certainty Factor method by using Visual Studio 2010 applications and using Microsoft Access 2013 as a database. Applications that are built can be an alternative in knowing the type of damage and provide alternative solutions in handling damage to PC computer hardware.

Photocopying machines are machines that make paper copies of documents and other visual images quickly and cheaply. PT. TRITANU has a copy machine product, such as the Fotocpy Sharp machine Type B201D. The rapid development of photocopying technology at this time can be seen that adult photocopying machines are now not only operating copier but can scan, print, and even fax. The rapid development of photocopying technology does not necessarily provide complete convenience to users. This can be seen from the high cost of consultation in the event of damage to the photocopy machine, requiring time to check the damaged machine, and not to mention other procedures in the repair that exist at PT. TRANANU. So from to provide solutions to these problems, the researchers made an Expert System for Handling Print Results Problems on Sharp Photocopying Machines Type B201D Using the Certainty Factor Method. System testing shows that the system is able to diagnose damage to the B201D copy machine based on the symptoms in the printout, even though these symptoms contain uncertainty. The results of the diagnosis are accompanied by Certainty Factor values that indicate the level of truth, accuracy of possible damage to the B201D copy machine.

Digital Dictionary of Computer and Network Engineering can be used in the learning process of computer and nerwork enginering students. Dictionaries are generally book-shaped is hard to carry because of their thick and heavy, now can be accessed anywhere by development of web technology. Digital dictionaries can be accessed through computers, laptops dan cellphones, this is make student easily in learning process. The digital dictionary can not only be used by students of computer and network engineering, but can also by general public, as long as you have an internet connection, anyone can be use this digital dictionaries. The main process in this digital dictionary is the search process. Binary search algorithm used in the search process of binary search algoritm. Binary search algoritm search is applied to word search in this digital dictionary, because this algoritm is intended for sequintal data. Program language that will be used to build this digital dictionary is PHP and MySQL.

Consumer purchasing patterns are a form of purchases made by consumers, whether someone or a lot of people to get the desired item by making a purchase transaction. One characteristic of the purchase pattern is the existence of acquiring something through exchanging money. This study aims to create an application that is used in determining consumer purchasing patterns by applying a priori algorithms and using Visual Basic 2010 as a tool for determining consumer purchase patterns. This application uses a priori algorithm calculation method where the sample consumer purchase data will be sorted and calculated by providing the value of the minium support and configuration parameters and based on the results of confidence the largest number of conclusions such as: can be used as information for determining sales, the application of a priori algorithms can provide information pattern combination item set from consumer purchase data that is with support above 15% and confidence above 50% on item set.

In entering competition the quality and predicate best the national and the quality of graduates, college trying to use information technology as an instrument for might win this competition. That IT can be used to fulfill and support the system in college, it needs assessment from information technology regularly. One method of conducting evaluation of the performance of the department of information technology is to use the framework Cobit 5 as yardstick efficiency in the use of information technology currently consisting of 5 (five) domain namely: EDM, APO, DSS, MEA A measuring instrument his as. With this method, researchers trying to make it a blueprint as the end result of utilization of information technology in college, so the management can plan exactly how of development of information technologies company to the next few years. An evaluation, known level capabilities inside the area mea and APO is entirely located at the level of 1 (performed) with the level of the target to be achieved is the level of 3 (established). Weakness governance information technology in STMIK Pelita Nusantara Medan is the lack of formalization rules and procedures management information technology. To reduce the gap between capability the current level was level and capability to be achieved, so STMIK Pelita Nusantara Medan must meet PA2.1, PA2.2, PA3.1 and PA3.2, that capability the current level was that is at the level of 1 could rise to the level of 3. Along with that, STMIK Pelita Nusantara Medan moved closer to the purpose..

Chicken disease is a condition in which the body organs of chickens cannot function normally. Diagnosing a disease requires symptoms that appear on the chicken’s body. To conduct a disease diagnosis, further examination is needed by specialists (experts). An expert system is a system that uses human knowledge that is entered into a computer to solve problems that are usually resolved by experts. The system created in solving problems uses the Bayes Theorem method in which the inference process uses Forward Chaining. The application was developed using the visual basic 2010 programming language and MySQL database.

The development of technology has been very significant, not only in the fields of information, industry, education, but in agriculture. Therefore technological sophistication is also utilized by corn farmers to obtain information about corn crops. Corn is included in plantation commodities in Indonesia. However, in corn cultivation there are various problems associated with the disease that is likely to attack the plant. Therefore, the authors designed the expert system of corn disease diagnosis using certainty method. This system will analyze based on symptoms - symptoms that are in the input by the user so that in obtaining a result of disease identification on corn in the name of the disease, the cause of the disease, the percentage value and how to handle it. This expert system is developed by using the visual basic application and Microsoft Access database with Certainty Factor method to measure the value of certainty of a hypothesis against a fact.