Information extraction in current Indian web documents

Prakash, Kolla Bhanu

doi:10.14419/ijet.v7i2.8.10332

Cited by 18 publications

(8 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As the training data is small, it may result in overfitting and less performance [11] for unseen test tuples. Hence, we apply the strategy of Data Augmentation which increases the diversity of data to train the neural networks.…”

Section: Data Augmentationmentioning

confidence: 99%

Automated Road Safety Surveillance System using Hybrid CNN-LSTM Approach

Donepudi¹

2020

IJATCSE

View full text Add to dashboard Cite

Among all the reasons for the occurrence of road accidents, the human state of drowsiness and Underage driving contributes a major share. With the rigid implementation of traffic rules and national schemes, it does not result in decreasing the accidents. Hence, there is a need for automation of surveillance which strictly restricts the teen driving and fatigue driving. In this paper, we introduce a face image descriptor-based combination of deep learning model i.e., convolution neural network (CNN) with ResNet50 architecture to predict age and a recurrent neural network (RNN) with LSTM architecture to detect the drowsiness in driver and alert them when they are drowsy. The algorithm is based on face recognition for age prediction and the blink frequency for detecting the fatigue. Image processing techniques are utilized to obtain the feature-based extraction for prediction. The proposed model developed could give a validation accuracy of 96% thus providing the promising results. This automation model thus could help the road safety authorities in their work and also decreases the occurrence of road accidents.

show abstract

Section: Data Augmentationmentioning

confidence: 99%

Automated Road Safety Surveillance System using Hybrid CNN-LSTM Approach

Donepudi¹

2020

IJATCSE

View full text Add to dashboard Cite

show abstract

“…[18] The model has trained info of speech found in English and wanted database languages. [19] These are compared through Multivariable linear regression classifier. [20] The speech is classified into frequency cepstrum coefficients and modulation spectral features.…”

Section: Emotion Recognitionmentioning

confidence: 99%

Emotion Recognition in Speech Processing using Fast Fourier Transform

Jammula¹,

gogireddy²,

Vege³

et al. 2019

IJRTE

View full text Add to dashboard Cite

The idea of acquiring the state of human emotions from one’s speech, we have gathered required data that makes one to understand the concept behind this process. Human emotions can be predicted by his/her facial expressions or by the tone of their voice. Reading the facial expressions is one of the major tasks involved in image processing. Likewise, each emotion holds different tone in one's voice. It requires a various emotional tone frequency to calculate and analyse the emotions. We need to fetch approximate frequencies of emotions. It’s the challenging task as each speaker has various pitches while speaking and frequencies of the same person varies in his emotion. Another main issue is the noise in the input while a person is speaking, due to less quality recordings or surrounding environment. List of basic emotions are Happy, angry, sad, bored, surprised, disgust, fear. For this project the prior important concept is speech recognition. The machine must be capable of reading the input in form as speech and must be capable of analysing various contents. The input given is converted into wav format. At the same time machine must be also capable of fetching the frequencies. The calculation is performed using many methodologies that are defined.

show abstract

“…The BAN gateway always feeds HAN taking into account the last requested amount of electricity, up until the HAN sends another message requesting a change. [23,24] Phase III. Billing Process The payment values are calculated by the BAN gateway for every HAN by multiplying the current electricity price with the electricity consumed, compile them with the former values, and the result is stored in the HAN network record in its database.…”

Section: B Price Changementioning

confidence: 99%

Lightweight Security and Privacy Safeguard strategies for Smart Grid management in Client-side Networks

Sri*,

Laditya,

Ram

et al. 2019

IJITEE

View full text Add to dashboard Cite

In order to improve the reliability and efficiency of the power grid, the smart grid uses different communication technologies. Smart grid allows bidirectional flow of electricity and information, about the state of the network and the preconditions of the clients, between the different parts of the network. Therefore, it reduces energy losses and generates and distributes electricity efficiently. Although smart grid improves the quality of network services, due to the nature of the power grid communication networks are exposed to cybersecurity threats along with the other threats. For example, electricity consumption messages sent by consumers to the utility through the wireless network can be captured, modified or reproduced by adversaries. As a consequence, the important challenges in smart grid seems to be security and privacy concerns. The smart grid update creates three main communication architectures: the first is communication between the utility companies and customers through diverse networks; that is, Local Area Networks (HAN), Construction Area Networks (BAN) and Neighboring Area Networks (NAN), we refer to these networks as client-side networks in our thesis. The second architecture is the communication through the vehicle-to-network (V2G) connection between the Electric Vehicles and the network to charge or discharge their batteries. The hindmost network is connection of the network with measurement units that extend throughout the network in order to monitor the status and send reports periodically to the main CC to estimate the status and detect erroneous data. The proposed schemes are promising solutions for the security and privacy problems of the three main communication networks in smart grid. The novelty of these proposed schemes is not only because they are robust and efficient security solutions, but also due to their lightweight communication and computing overhead, which qualifies them to be applicable in devices with limited capacity in the network. Therefore, this work is considered an important progress towards a more reliable and authentic intelligent network.

show abstract

Information extraction in current Indian web documents

Cited by 18 publications

References 6 publications

Automated Road Safety Surveillance System using Hybrid CNN-LSTM Approach

Automated Road Safety Surveillance System using Hybrid CNN-LSTM Approach

Emotion Recognition in Speech Processing using Fast Fourier Transform

Lightweight Security and Privacy Safeguard strategies for Smart Grid management in Client-side Networks

Contact Info

Product

Resources

About