Browse Articles
Discover research articles across all indexed journals
MOVIE RECOMMENDATION BASED ON HUMAN EMOTION USING CNN
Emotion plays a vital role in human’s action. Humans highly influenced by movie and music they listen and watch. During this pandemic time, video has become popular and dominant medium and especially OTT (Over the top) service has seen an exponential a growing activity and many companies are now focusing on trying to focus on this field. In past few years, there is much improvement of new techniques to gather real time emotions. In this Paper, Tried to provide the System design model by which recommendation system can be improved on the basis of human emotions to provide dynamic content recommendation to the user. By capturing real-time emotion using deep neural network technique and combining traditional Content- based recommendation we will recommend the list of movies to user. For this implementation we used python to demonstrate its working. Our approach is trying to overcome problem in existing method.
AN IOT BASED CHARGING MODULE FOR RECHARGABLE BATTERIES OF ELECTRIC VEHICLES
Fossil fuels are a non-renewable resource that is exhaustible in near future. An alternative for such automobiles is using electricity which can be generated by renewable resources to power up the vehicle. This paper gives an idea about charging these Electric-Vehicles (e-Vehicles) using a renewable energy source such as solar using solar panels and provide the maximum power generated. The proposed system is controlled using the Arduino UNO to view the amount of charge generated through an LCD. GSM modules give alert messages if there is an overcharge or undercharges. A rain sensor is also integrated to alert the user if there are any changes in climatic conditions that affect the charging of PV cells. The entire status of the battery along with the amount of power being transferred is uploaded to the cloud and can be seen through a webpage at any time. The main aim of this paper is in reducing the consumption of non-renewable energy sources and greenhouse gas emissions reduction.
GUITAR SIDE BODY MAKING TECHNIQUE USING SCREW PRESS AND HEATER
The guitar musical instrument industry located in Baki District, Sukoharjo, Central Java, Indonesia, is still using the manual method of guitar manufacturing process currently. The problem is that the guitar production time is not only 2 hours and is considered a long time and non-uniformity of the quality. A better Technology is needed to help improve the quality and quantity of guitar production. The purpose of this research is to design and manufacture a guitar side body press with a heater using an electric element and a gas stove. The benefit of this tool is that it speeds up the process of forming the side body of the guitar. The research method begins with a preliminary study, design, manufacture, assembly, and testing the performance of the guitar side body-forming machine. The research resulted in a unit of guitar side body pressing machine with a length of 1200 (mm) Width 520 (mm) Height 1000 (mm). 2000 (watt) electric heater. The pressing process uses a manual screw press. Testing showed that the best performance of the guitar side body press machine is at a temperature of 75℃ with 5 minutes pressing time. The machine produced by this research can be used by guitar SMEs to improve quality, productivity, and competitiveness.
CFD ANALYSIS OF SOLAR OPERATED EARTHEN POTAIR COOLER
The present air-ventilation methods are evaporative coolers, air conditioning, fans. But running these products need high-rate electricity. The production of electricity is ultimately responsible for hot and humid conditions i.e., global warming. In hot and humid conditions, the need to feel relaxed and comfortable has become one of few needs and for this purpose utilization of systems like air-conditioning and refrigeration has increased rapidly. These systems are most of the time not suitable for villages due to longer power cut duration and the high cost of products. Solar power systems being considered as one of the paths towards more sustainable energy systems, considering solar cooling systems in villages would comprise of many attractive features. This technology can efficiently serve large latent loads and greatly improve indoor air quality by allowing more ventilation while tightly controlling humidity. Despite increasing performance and mandatory energy efficiency requirements, peak electricity demand is growing and there is currently no prevalent solar air-cooling technology suited to residential. This research CFD analysis of solar operated earthen pot air cooler for domestic applications. Velocity airflow within the Earthen Pot in that air velocity has been increased with reducing the pressure results the temperature variation. In Temperature contour the temperature rate reduces with 8 to 10 OC of room temperature and the analysis gives the temperature difference and air velocity variations.
ECU EXTRACT ANALYZER AND MERGER TOOL
AUTOSAR XML (.arxml) is a format introduced by the Automotive Open System Architecture consortium to contain the data used in and required by Electric Control Units which is based on Automotive Open System Architecture. Electric Control Unit development is done by 3 groups i.e. OEM like BMW, Tier1like Bosch, and Tier2 like a vector. Each one has its own data that is required for Software development. OEM shares the information with Tier1.Tier1 shares the same with Tier2 for implementing into Electric Control Unit. The DataBase Container (DBC) describes the properties of the Controller Area Network (CAN) / Controllable Area Network Flexible Data Rate (CAN FD). Data analysis in ECU Extract ARXML is very difficult due to the complex structure of data. Performing two-way and three-way comparisons/merges of XML files is difficult. To handle the mentioned challenges manually is a big task that requires huge manual efforts and time which may even result in a manual error. In this paper, we are presenting a newly developed Python-based tool that overcomes the above-mentioned drawbacks.
Batch Comparative Adsorption Studies of Fluoride Using Activated Carbon Derived from Magnifera Indica (Mango) Seeds and Peltophorum Pterocarpum (Copper Pod) Seeds
Activated Carbon (AC) is frequently used for purification due to its ability to remove pollutants either from air or water. Its porous structure allowed it to capture pollutants and is widely recommended for numerous applications in water and wastewater treatment. However, AC applications fields are restricted due to high costs. In this project, AC was prepared from locally available Magnifera Indica (Mango) seeds and Peltophorum Pterocarpum(Copper Pod) seeds. The removal of Fluoride from the synthetic sample by adsorption on physically activated carbon prepared from Magnifera Indica (Mango) seeds and Peltophorum Pterocarpum(Copper Pod) seeds have been carried out at room temperature 32 ± 1oC. Batch adsorption processes were conducted for the removal of Fluoride on physically activated carbon for optimum time, optimum dosage, optimum pH sorption kinetics sorption equilibrium. The optimum contact time, adsorbent dose, and pH for carbon was 35 min, 100 mg and 8.00 and 40min, 160mg and 8.50 respectively, with a fluoride removal efficiency of 78.96%, 96%, 97% and 45%, 58%, 96% respectively have been observed. Adsorption kinetics and sorption isotherm models were studied by altering different process parameters like adsorption time and dosage respectively. Langmuir, Freundlich, and Temkin isotherms are studied through this experimental data and found to be well followed.
MASK R-CNN FOR FIRE DETECTION
Object detection has an increasing amount of attention in recent years due to its wide range of applications and recent technological breakthroughs. Deep learning is the state-of-art method to perform object detection. This task is under extensive investigation in both academics and real-world applications such as security monitoring, autonomous driving, transportation surveillance, drone scene analysis, robotic vision, etc., It is a computer technology related to computer vision and image processing that deals with detecting instances of semantic objects of a certain class (such as humans, buildings, or cars) in digital images or videos. It not only provides the classes of the objects in an image but also localizes them in that particular image. The location is given in the form of bounding boxes or centroids. Instance segmentation may be defined as the technique that gives fine inference separately for each object by predicting labels for every pixel of that object in the input image. Each pixel is labeled according to the object class within which it is enclosed. We deal with Mask Region-Based Convolutional Neural Network (Mask R-CNN) to implement instance segmentation and detection of fire in a video or an image which can be used in real-world such as automatic fire extinguisher and alert systems. The training was done using Mask R-CNN for object detection with ResNet-101 backbone, with a 0.001 learning rate and 2 images per GPU. With this, the proposed framework can detect fire using Mask Region-Based Convolutional Neural Network and can send immediate alert to the user if fire is detected
HEAD MOTION DETECTION
The proposed system is a system that can be used for surveillance and monitoring applications. The development of an efficient real-time video head detection system is motivated by its potential for deployment in the areas where security is the main concern. The proposed system presents a platform for real-time video head detection and subsequent generation of an alarm condition as soon as the human head is detected. The prototype consists of a platform mounted with a camera that provides continuous feedback of the online exam environment.
BITCOIN PRICE PREDICTION
Bitcoin is currently the leading global provider of cryptocurrency which has recently received a lot of attention from the public. In recent years, the Bitcoin network has attracted investors, businesses, and corporations while facilitating services and product deals. The factors that affect the Bitcoin price and the patterns behind its fluctuations can be predicted by using various machine learning models like LSTM, ARIMA, and Facebook prophet. This prediction may give better insights about the bitcoin price to the people who are investing in the bitcoin. Accurate prediction of the future trend in the closed price of the day for a given cryptocurrency is a challenging task. Machine Learning models can be used to predict the closed price for a given cryptocurrency.
DYNAMIC ANALYSIS AND DESIGN OF MULTISTOREYED R.C STRUCTURAL FRAME WITH PILE FOUNDATION
The structure is presumed to be situated in seismic Zones II and V. The mathematical building model (F.E.M Model) of Whole building has been modelled in ETABS 16.2 All the beams and columns are modelled as frame elements. All the slabs are modelled as thin shell elements. To ascertain the seismic load on each floor level of whole structure two types of seismic analysis procedure have been considered as per IS1893-2002 and 2016. Equivalent static approach (linear Static method) and Response spectrum method (Linear Dynamic Method). Parameters considered to understand the seismic response of the multi-storeyed buildings are, Modal Mass participation factor, Modal time period, Lateral displacement, Storey drift, Storey shear of the structure. The whole structural systems are strengthen with vertical shear walls. SAFE (Slab analysis by Finite Element Method) is a software essentially made for analyse and designs of the slabs, since Pile Raft is also a slab therefore SAFE is the ultimate tool for it. Pile Caps have modelled as Slab with Thick plate Action, for base reactions Link between ETABS and SAFE have been used. The parameters consider for finalise the designs of piles and pile caps are, vertical settlement, Punching Shear stress under columns and shear walls and for piles as well. Flexural reinforcements are tabulated in both Longitudinal and transvers direction for the both the faces i.e. Top face and bottom. After all the study a comparison is made between the structure situated in Zone II and Zone V We had observed that when the seismic zones changes and the intensity of seismic forces increases, the design of foundation considerably changes and there is tremendous amount of increase in steel.
COMPARATIVE ANALYSIS AND DESIGN OF SOLID AND MULTILAYERED COMPOSITE PRESSURE VESSEL
A pressure vessel designed to work under a high internal pressure condition, so selecting material and the vessel's design are the most crucial parts. This paper compares the conventional (solid) pressure vessel made of a steel s515-gr70 and composite material pressure vessel made of epoxy carbon. The static deformation and von-mises stress for a pressure vessel were calculated and compared with composite and multi-layered composite material. The main objective of this paper is to compare the conventional (solid) pressure vessel with composite material pressure vessel and multi-layered composite pressure vessel and find out the best suitable material and design for the pressure vessel. To reduce the stress on the object, we are designing the multilayer vessel and calculating the deformation and stress-induced in the vessel. All these values compare with all other material combinations and conclude which composite is the most suitable and efficient pressure vessel.
NATURAL LANGUAGE PROCESSING AND ITS EFFECTS ON SOCIETY AS A WHOLE: A SURVEY
Over the past 25 years, computational linguistics has grown to become both an interesting area of scientific research and upcoming technology that is increasingly being put into consumer products (for example, applications such as Apple’s Siri, Microsoft Cortana or Skype). There are main four key factors that are the driving force behind these developments: (i) an immense increase in computing power, (ii) the availability of huge amounts of linguistic data, (iii) the development of highly effective machine learning (ML) methods, and (iv)a more enriched understanding of the structure of human language and its application in social contexts.[2] Natural language processing (NLP) is the branch of Artificial Intelligence (AI) that deals with communication: How can a computer be programmed to understand, process, and generate language just like a human being? With the increasing use of the Web and other text application areas, there has been an increasing demand in both text mining and natural language processing (NLP). The main goal of Natural Language Processing (NLP) is to implement within computers the skill set required to understand a normal human language or natural language. It is related to the field of HCI ie. Human-Computer Interaction. One of the motivations of NLP is for the society whose access to web information is blocked simply by their inability to use the keyboard and OS. Natural language comes under the purview of AI with the goal of understanding and recreating meaningful expressions in human language. AI is the ability of a machine to imitate intelligent human behavior. This paper presents a survey on Natural language processing (NLP) and its effects on society as a whole. Here I talk about sentiment analysis which has a long-term effect on society. I also talk about social media mining which results in a huge amount of structured social information which is extremely important in the predictive analytics industry also. Then I discuss the future scope of NLP in the society.
COGNITIVE NATURAL LANGUAGE PROCESSING ASSISTANT FOR POST-OPERATIVE CARE
The vast intricacies that hover over post-operative care often lead to higher risk and inherent damage, when ignored. Postoperative care has several complications, ranging from decreased quality of care due to enormous patient to doctor gap ratio to unavailability of remote healthcare. To ensure that these issues are eradicated, the Cognitive NLP assistant using Dialogflow is proposed. This assistant will hold a regular conversation similar to what a medical professional would, but at the feasibility of the user’s home. It will monitor and track symptoms. The assistant will suggest prompts, based on prior charts and care suggested for the respective surgery. The proposed methodology is achieved by using the Rule-Based Grammar algorithm and the ML matching algorithm. The assistant’s services will extend to connecting with the medical professional if any abnormalities arise and cover communication gaps, by providing the details it has gathered. This assistant has reduced the burden of the physician while providing care to the user and eliminating the need to travel distances to seek treatment for non-emergent issues along with reducing the stress to schedule an appointment. It will function as a whole care provider by being available around the clock to tend to the user.
STABILITY INDICATING METHOD DEVELOPMENT AND VALIDATION FOR SIMULTANEOUS ESTIMATION OF DEXTROMETHORPHAN HBr, PHENYLEPHRINE HCl AND CHLORPHENIRAMINE MALEATE IN THEIR COMBINED SYRUP DOSAGE FORM BY REVERSE PHASE HIGH PERFORMANCE LIQUID CHROMATOGRAPHY
The present work describes a validated reverse phase high performance liquid chromatographic method for simultaneous estimation of Dextromethorphan HBr, Phenylephrine HCl and Chlorpheniramine Maleate in syrup dosage form. The quantification was carried out using Phenomenex Luna 5µm C8(2)100Å;250×4.6mm (Part No.00G-4249-E0) column and mobile phase comprised of Buffer pH 3.2 and Acetonitrile with gradient elution mode. The flow rate was 1.0 ml/min and the eluent was monitored at 220 nm. The selected chromatographic conditions were found to effectively separate Dextromethorphan HBr, Phenylephrine HCl and Chlorpheniramine Maleate were about 11.1 min, 3.7 min and 10.3 min respectively. Linearity were found to be in the range of 80-120 μg/ml, 40-60 μg/ml and 16-24 μg/ml for Dextromethorphan HBr, Phenylephrine HCl and Chlorpheniramine Maleate respectively. The percentage recoveries of all the drugs were found to be 100.5-101.8%, 99.2-101.8% and 100.1-101.8% for Dextromethorphan HBr, Phenylephrine HCl and Chlorpheniramine Maleate. The proposed method was found to be fast, specific, accurate, precise, and reproducible and can be used for simultaneous estimation of these drugs in syrup formulation.
DESIGN & DEVELOPMENT OF A BIG DATA ALGORITHM OPTIMIZATION TECHNIQUE FOR A SALES SIMULATION SYSTEM OF A BUSINESS ORGANISATION
When sales representatives and customers negotiate, it should be confirmed that the ultimate deals can render a high enough professional t for the mercantilism company. Massive corporations have completely different strategies of doing this, one amongst that is to run sales simulations. Such simulation systems typically have to be compelled to perform complicated calculations over massive amounts of information that successively needs economical models and algorithms. This research paper intends to judge whether or not it's potential to optimize Associate in Nursing extend an existing sales system known as per centum, that is presently laid low with intolerably high running times in its simulation method. this can be done through analysis of this implementation, followed by improvement of its models and development of economical algorithms. The performance of those optimized and extended models area unit compared to the present one so as to judge their improvement. The conclusion of this research work is that the simulation method in per centum will so be optimized and extended. The optimized models function as a symptom of thought that shows that results just like the first systems are often calculated inside < 1 Chronicles of the first time period for the most important range of shoppers.
PREDICTIVE ANALYSIS OF CLASSIFICATION ALGORITHMS ON BANKING DATA
The entire human activity is generating about 2.5 quintillion bytes of data per day. It is growing exponentially. The tremendous volume of data is getting generated by different banking services, including real-estate loan, deposit accounts services, credit card issuing service, commercial loan lending and many more. The bank data is used for business analytic that keeps eyes on past growth also Predictive analysis opens doors for future growth. For this project, during the predictive analysis, we used some historical data generated by an organization during loan processing. We intended to predict whether a new applicant will have granted a loan or not. The analysis was done by building Machine Learning (ML) models using different classification algorithms like Random Forest(RF), Logistic Regression(LR), Support Vector Machine(SVM), Decision Tree(DT), Gradient Boosting, XG Boost. The classifier models were trained and tested on hyper parameters as the performance of each model was analyzed using Confusion Matrix. We compared efficiencies using Recall, Precision, Accuracy and F1 score in a tabular form among all models. LR yields an F1 score of 0.90 and 0.62 for class 1 (approved) and class 0 (not Approved), respectively. Also, XG Boost gave F1 score of 0.89 and 0.64 for the same class 1 and 0.
A PROFICIENT PILOT DESIGNING PATTERN FOR CHANNEL ESTIMATION IN MIMO-OFDM SYSTEMS
The high data transmission rates in the communication channel is a recent growing demand which is achieved by using the multiple input multiple output-orthogonal frequency division multiplexing (MIMO-OFDM) systems because of their unique properties.. The OFDM systems that adopt the cyclic-delay diversity (CDD) scheme require only one inverse discrete Fourier transform (IDFT) operation at the transmitter. Hence, the CDD provides a low-complexity by means of increasing the transmission diversity in MIMO-OFDM systems. However, the optimal pilot sequences which minimize the mean square error (MSE) of the channel estimate in traditional MIMO-OFDM systems are inapplicable to CDD-OFDM systems. Thus this paper originates by deriving the criteria which yield the minimum MSE of both the least square (LS) channel estimate and the MMSE channel estimate in CDD-OFDM systems. The derived criteria are then used to develop a general methodology for determining the optimal pilot sequence. Significantly, the proposed design methodology enables the status of the channel to be estimated using single OFDM symbol. We have compared the performances of channel estimation algorithm by measuring bit error rate vs. SNR with BPSK, QPSK 16-PSK and 256-PSK modulation schemes.
THE CULTURE OF ILMU AMALI AMAL ILMI IN GROWING ULUL ALBAB CHARACTERS OF SANTRI
This study aims to determine the pattern of cultural development of the culture of ilmu amaly amal ilmy in cultivating the ulul albab character of the students. The research method uses descriptive qualitative research. Place and robject research in Pondok Pesantren Raudlatul Ulum Guyangan Pati. The results showed that the pattern of cultural development ilmu amali amal ilmi are time management, educational punishment and an integrative curriculum can fosters the character of ulul albab, namely discipline, independence and competence in the fields of general science and religion.
SIMULATING A REAL TIME INTERACTIVE TELEMEDICINE IN A TYPICAL NIGERIAN HOSPITAL
Telemedicine involves the provision of health care and sharing of medical knowledge using telecommunications technologies. Diagnostic, patient education health can be provided via telemedicine. Due to the shortage of medical experts, hospitals coupled with geographic isolation and lack of opportunities, delay in the system, and cost of transportation, it important to develop a robust system that will improve and change the present state of health care and delivery services. In this research work, interactive service telemedicine is designed to provide tools for overcoming the unequal distribution of health care. The proposed system also provides opportunities for patients to gain easy access to medical personnel. The methodology used for this research work is the Dynamic System Development Method (DSDM). As a result of this research medical personnel, have real homes. Telemedicine is the future of health care.
DEEP LEARNING FOR MOLECULAR PROPERTY PREDICTION OF NOVEL COMPOUND
Images of Molecules are broadly used to foresee atom properties in educating and synthetic exploration. A prepared physicist can undoubtedly infer particle properties by dissecting its construction and assess its useful gatherings. To foresee, for instance, the water dissolvability of a natural compound a scientist would instinctively tally the quantity of polar gatherings, think about the size of the molecule, and gauge the water/atom communication by tallying the quantity of H-security benefactors and acceptors. Subsequently, 2D molecule portrayals and their straightforwardly available highlights ought to give sufficient data to anticipate the atom's design subordinate properties. To help this postulation, image based AI approach convolutional neural network, clustering and transfer learning are utilized and assessed in this work. At long last properties of new molecule produced utilizing deep learning were anticipated.