Deep learning methods for Fake News detection

Maslej-Krešňáková, Viera; Sarnovský, Martin; Butka, Peter

doi:10.1109/cinti-macro49179.2019.9105317

Cited by 24 publications

(7 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The convolutional layers were obtained from the matrix of generated sequences and binary vectors representing the selected small target molecules. The output of the convolutional layers was used as input of the pooling layer with the aim of down-sampling the features learned by the filters [ 76 ]. The output of the convolutional and pooling layers was fed to the Fully Connected (FC) layers [ 77 , 78 ].…”

Section: Methodsmentioning

confidence: 99%

Smart Approach for the Design of Highly Selective Aptamer-Based Biosensors

et al. 2022

View full text Add to dashboard Cite

Aptamers are chemically synthesized single-stranded DNA or RNA oligonucleotides widely used nowadays in sensors and nanoscale devices as highly sensitive biorecognition elements. With proper design, aptamers are able to bind to a specific target molecule with high selectivity. To date, the systematic evolution of ligands by exponential enrichment (SELEX) process is employed to isolate aptamers. Nevertheless, this method requires complex and time-consuming procedures. In silico methods comprising machine learning models have been recently proposed to reduce the time and cost of aptamer design. In this work, we present a new in silico approach allowing the generation of highly sensitive and selective RNA aptamers towards a specific target, here represented by ammonium dissolved in water. By using machine learning and bioinformatics tools, a rational design of aptamers is demonstrated. This “smart” SELEX method is experimentally proved by choosing the best five aptamer candidates obtained from the design process and applying them as functional elements in an electrochemical sensor to detect, as the target molecule, ammonium at different concentrations. We observed that the use of five different aptamers leads to a significant difference in the sensor’s response. This can be explained by considering the aptamers’ conformational change due to their interaction with the target molecule. We studied these conformational changes using a molecular dynamics simulation and suggested a possible explanation of the experimental observations. Finally, electrochemical measurements exposing the same sensors to different molecules were used to confirm the high selectivity of the designed aptamers. The proposed in silico SELEX approach can potentially reduce the cost and the time needed to identify the aptamers and potentially be applied to any target molecule.

show abstract

Section: Methodsmentioning

confidence: 99%

Smart Approach for the Design of Highly Selective Aptamer-Based Biosensors

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Neural networks 28,29 , especially feedforward networks, are popular for text classification. They require less training data compared to advanced models like CNN and LSTM.…”

Section: Neural Network Model With Feed-forward Architecturementioning

confidence: 99%

Research on emotional interaction and user cognition

Ying,

Zhang,

Huang

et al. 2023

Fourth International Conference on Computer Science and Communication Technology (ICCSCT 2023)

View full text Add to dashboard Cite

This paper focuses on the field of emotional interaction, specifically using text data to evaluate expressed emotions. It aims to assess the emotional state of humans in order to determine how robots should adjust their behavior. To detect emotions in text data, various methods and algorithms can be utilized. In this study, the authors combine dictionary methods with machine learning algorithms for emotion detection. However, emotion labels are often vague and subjective, leading to a multi-label problem. To address this, the paper conducts experiments using naive Bayes, support vector machine, and neural network classifiers. Although the accuracy of emotion classification is low, the study demonstrates the importance of specific words in expressing emotions.

show abstract

“…Through said process, words are mapped and stored in real number vectors of discrete or distributed representation for further use [19]. This preprocessing step can be done automatically by using algorithms such as Word2Vec and Google News [17,29].…”

Section: Deep Learning Applicationsmentioning

confidence: 99%

“…Convolutional Neural Network (CNN). These are a type of neural networks used for data processing, which get their name from the mathematical operation known as convolution [17]. This type of algorithms are based on the human visual cortex [36], which allowed them to gain popularity in the area of natural language processing [14].…”

Section: Deep Learning Algorithmsmentioning

confidence: 99%

See 1 more Smart Citation

Fake News Detection in Internet Using Deep Learning: A Review

Barreto

Córdova

Arana

2021

Studies in Computational Intelligence

View full text Add to dashboard Cite

The main objective of this research was to explore, from a reflexivity approach, the current state of Deep learning techniques for automatic detection of fake news on the Internet, analyzing the most important Deep learning algorithms and studies on their effectiveness in detecting distrustful information. The research methodology employed was bibliographic, documentary and descriptive. The information was collected from several scientific articles provided by indexed journals and web platforms, using keywords such as "fake news", "Deep learning" and "neural networks" for the compilation. As a result of this research, it was concluded that Deep learning techniques present a better performance than conventional methods and will be of great importance in the future of war against fake news due to their potential in automatic detection.

show abstract

Deep learning methods for Fake News detection

Cited by 24 publications

References 14 publications

Smart Approach for the Design of Highly Selective Aptamer-Based Biosensors

Smart Approach for the Design of Highly Selective Aptamer-Based Biosensors

Research on emotional interaction and user cognition

Fake News Detection in Internet Using Deep Learning: A Review

Contact Info

Product

Resources

About