March 2020 - Journal of Artificial Intelligence and Capsule Networks
Latest Publications


TOTAL DOCUMENTS

56
(FIVE YEARS 56)

H-INDEX

7
(FIVE YEARS 7)

Published By Inventive Research Organization

2582-2012
Updated Saturday, 24 July 2021

Author(s):  
Milan Tripathi

The government's months-long total lockdown in response to the COVID19 outbreak has resulted in a lack of physical connection with others. This resulted in a massive increase in social media communication. Twitter has become one of the most popular places for people to communicate their thoughts and opinions. As a result, massive amounts of data are created every day. These data can assist businesses in making better judgments. In the case of Nepal, there has been relatively little investigation into the text's analysis. Because few researchers are working in the field, development is slow. In this study, Four language-based models for sentiment analysis of Nepali covid19 tweets are designed and evaluated. Because the number of individuals using social media is expected to skyrocket in the next few days, companies will benefit from an AI-based sentiment analysis system. It will greatly assist firms in adapting to the changing climate.


Author(s):  
Prachu J. Patil ◽  
Ritika V. Zalke ◽  
Kalyani R. Tumasare ◽  
Bhavana A. Shiwankar ◽  
Shivani R. Singh ◽  
...  

One of the many challenges that the world faces is traffic hazard. The major cause of this traffic risk is the presence of a huge number of vehicles on the road. As a result, it generates the most challenging issues, leading to an increase in the death toll due to road accidents that occur throughout the world. As a result, it necessitates the need to provide adequate transportation facilities, which will reduce the number of collisions and save human lives. The GPS, GSM, accelerometer, Arduino UNO technology, and vibration sensor are used to design and develop a vehicle accident detection model. The proposed approach is classified into three stages to prevent and detect the vehicular accidents. At the detection stage, a vibration sensor will be utilized to determine the position of the accident and to alert the user by sending SMS via the GSM module, which will include the user's data stored in Android applications. This data will be taken from the GPS module. The second phase occurs when moderate accidents occur and in such situation, the location will be detected by using a GPS module. After that, the nearby hospital receives a message about the accidents and accordingly they provide services to the accidents. At the same time, after detecting the location, a patient receives a message from the hospital urging them to take precautions. .


Author(s):  
Samuel Manoharan J

Herbal plants are crucial to human existence for medical reasons, and they can also provide free oxygen to the environment. Many herbal plants are rich in therapeutic goods and also it includes the active elements that will benefit future generations. Many valuable plant species are being extinguished and destroyed as a result of factors such as global warming, population growth, occupational secrecy, a lack of government support for research, and a lack of knowledge about therapeutic plants. Due to the lag of dimensional factors such as length and width, many existing algorithms fail to recognize herbal leaf in all seasons with the maximum accuracy. Henceforth, the proposed algorithm focuses on the incomplete problems in the datasets in order to improve the detection rate for herbal leaf identification. The inclusions of dimension factors in the datasets are performing good results in the image segmentation process. The obtained result has been validated with a machine learning classifier when combined with ex-or gate operation is called deep knowledge-based identification. This two-stage authentication (TSA) procedure is improving the recognition rate required for the detection of herbal leaf. This fusion of image segmentation with machine learning is providing good robustness for the proposed architecture. Besides, intelligent selection of image segmentation techniques to segment the leaf from the image is improving the detection accuracy. This procedure is addressing and answering the drawbacks associated with the detection of the herbal leaf by using many Machine Learning (ML) approaches. Also, it improves the rate of detection and minimizes the classification error. From the results, it is evident that the proposed method has obtained better accuracy and other performance measures.


Author(s):  
Joy Iong-Zong Chen ◽  
Kong-Long Lai

With the exponential increase in the usage of the internet, numerous organisations, including the financial industry, have operationalized online services. The massive financial losses occur as a result of the global growth in financial fraud. Henceforth, devising advanced financial fraud detection systems can actively detect the risks such as illegal transactions and irregular attacks. Over the recent years, these issues are tackled to a larger extent by means of data mining and machine learning techniques. However, in terms of unknown attack pattern identification, big data analytics and speed computation, several improvements must be performed in these techniques. The Deep Convolution Neural Network (DCNN) scheme based financial fraud detection scheme using deep learning algorithm is proposed in this paper. When large volume of data is involved, the detection accuracy can be enhanced by using this technique. The existing machine learning models, auto-encoder model and other deep learning models are compared with the proposed model to evaluate the performance by using a real-time credit card fraud dataset. Over a time duration of 45 seconds, a detection accuracy of 99% has been obtained by using the proposed model as observed in the experimental results.


Author(s):  
Vivekanadam Balasubramaniam

We present a complete overview of routing protocols, routing algorithms, path planning, and cloud deployment for vehicle navigation in several fields of study in this article. In this article, we compare several approaches and algorithms with the goal of identifying the best feasible ones based on the type of application being utilized. In general, navigation of vehicles will be based on models and methods. Hence in this paper each characteristics are examined in detail and the research has been done accordingly. Under each characteristic, performance evaluation criteria are separately analysed. Questions are also provided for which the literature review serves as a form of discussion, according to the research challenge and criteria. For path planning, node-based as well as traditional algorithms are considered as the best choices. Similarly, the performance is significantly improved when using hybrid routing protocols and route planning methodologies that prefer graph based techniques. It has been observed that, a number of future research directions such as routing algorithm with queuing theory and path planning with critical link methods also serve the probable domains. This work is a concise comprehensive study of the various characteristics of a vehicle with respect to navigation. A comparison of techniques, algorithms and methods by using the standard performance criteria has also been elaborated.


Author(s):  
Smys S ◽  
Haoxiang Wang

The concept of interconnecting smart vehicles and advancements in automotive automation leads to beneficial outcomes, such as a reduction in road fatalities and congestion. However, including a chain of automation in the attack surface will expand the attack surface and expose the security of automobiles to malicious infiltration. The proposed methodology provides access to specific users while restricting the third party requests. Moreover, it also makes use of data exchange that takes place between the roadside units and vehicle to track the vehicle status without compromising the in-vehicle network. To ensure a valid and authentic communication, vehicles with a proper and verifiable record will only be allowed to exchange messages in the blockchain network. Using qualitative arguments, we have identified that the proposed work is resilient to identified attacks. Similarly, quantitative experimentation indicates that this methodology shows a storage size compatibility and suitable response time in realistic scenarios. Simulation results indicate that, the proposed work shows positive results to secure vehicular networks, vehicular forensics and trust management.


Author(s):  
Thivaharan S ◽  
Srivatsun G

With the use of Ecommerce, Industry 4.0 is being effectively used in online product-based commercial transactions. An effort has been made in this article to extract positive and negative sentiments from Amazon review datasets. This will give an upper hold to the purchaser to decide upon a particular product, without considering the manual rating given in the reviews. Even the number words in an inherent positive review exceeds by one, where the present classifiers misclassify them under negative category. This article addresses the aforementioned issue by using LSTM (Long-Short-Term-Memory) model, as LSTM model has a feedback mechanism based progression unlike the other classifiers, which are dependent on feed-forward mechanism. For achieving better classification accuracy, the dataset is initially processed and a total of 100239 short and 411313 long reviews have been obtained. With the appropriate Epoch iterations, it is observed that, this proposed model has gain the ability to classify with 89% accuracy, while maintaining a non-bias between the train and test datasets. The entire model is deployed in TensorFlow2.1.0 platform by using the Keras framework and python 3.6.0.


Author(s):  
Jeena Jacob I. ◽  
Ebby Darney P.

The throughput of wireless multi-channel networks are enhanced using artificial intelligence algorithm. The performance of the network may be improved while reducing the interference. This technique involves three steps namely creation of wireless environment specific model, performance optimization using the right tools and improvement of routing by selecting the performance indicators cautiously. Artificial bee colony optimization algorithm and its evaluative features positively affects communication in wireless networks. The simple behavior of bee agents in this algorithm assist in making synchronous and decentralized routing decisions. The advantages of this algorithm is evident from the MATLAB simulations. The nature inspired routing algorithm offers improved performance when compared to the existing state-of-the-art models. The simple agent model can improve the performance values of the network. The breadth first search variant is utilized for discovery and deterministic evaluation of multiple-paths in the network increasing the overall routing protocol output.


Export Citation Format

Share Document