A comparative study of deep learning based language representation learning models

Self Cite

Sentiment analysis (SA) is widely used today in many areas such as crime detection (security intelligence) to detect potential security threats in realtime using social media platforms such as Twitter. The most promising techniques in sentiment analysis are those of deep learning (DL), particularly bidirectional encoder representations from transformers (BERT) in the field of natural language processing (NLP). However, employing the BERT algorithm to detect crimes requires a crime dataset labeled by the lexiconbased approach. In this paper, we used a hybrid approach that combines both lexicon-based and deep learning, with BERT as the DL model. We employed the lexicon-based approach to label our Twitter dataset with a set of normal and crime-related lexicons; then, we used the obtained labeled dataset to train our BERT model. The experimental results show that our hybrid technique outperforms existing approaches in several metrics, with 94.91% and 94.92% in accuracy and F1-score respectively.

Section: Related Workmentioning

confidence: 99%

Crime prediction using a hybrid sentiment analysis approach based on the bidirectional encoder representations from transformers

Boukabous¹,

Azizi²

2022

Self Cite

“…TL is a machine and deep learning research area that aims to transfer knowledge from one or more source tasks to one or more target tasks [15]. Supposing a source domain , a learning task , a target domain , and a learning task ; TL serves in improving the learning of the target predictive function (. )…”

Section: Transfer Learningmentioning

confidence: 99%

Accelerating the update of a DL-based IDS for IoT using deep transfer learning

Idrissi¹,

Azizi²,

Moussaoui³

2021

Self Cite

<p>Deep learning (DL) models are nowadays broadly applied and have shown outstanding performance in a variety of fields, including our focus topic of "IoTcybersecurity". Deep learning-based intrusion detection system (DL-IDS) models are more fixated and depended on the trained dataset. This poses a problem for these DL-IDS, especially with the known mutation and behavior changes of attacks, which can render them undetected. As a result, the DL-IDShas become outdated. In this work, we present a solution for updating DL-ID Semploying a transfer learning technique that allows us to retrain and fine-tune pre-trained models on small datasets with new attack behaviors. In our experiments, we built CNN-based IDS on the Bot-IoT dataset and updated it on small data from a new dataset named TON-IoT. We obtained promising results in multiple metrics regarding the detection rate and the training between the initial training for the original model and the updated one, in the matter of detecting new attacks behaviors and improving the detection rate for some classes by overcoming the lack of their labeled data.</p>

“…First, we define the architecture and strategy of every method then critically examine their main strengths and limitations. Secondly, we train a text encoder, char-convolutional neural network recurrent neural network (CNN-RNN) text embeddings [8], on a dataset of text captions to read the sentences and extract the relevant attributes. Then, we give a full implementation and adaptation on bird [9] and common object in context (COCO) [10] datasets, we compare their efficiencies on multiple levels like generated Indonesian J Elec Eng & Comp Sci ISSN: 2502-4752 …”

Section: Introductionmentioning

confidence: 99%

Optimal text-to-image synthesis model for generating portrait images using generative adversarial network techniques

Berrahal¹,

Azizi²

2022

Self Cite

The advancements in artificial intelligence research, particularly in computer vision, have led to the development of previously unimaginable applications, such as generating new contents based on text description. In our work we focused on the text-to-image synthesis applications (TIS) field, to transform descriptive sentences into a real image. To tackle this issue, we use unsupervised deep learning networks that can generate high quality images from text descriptions, provided by eyewitnesses to assist law enforcement in their investigations, for the purpose of generating probable human faces. We analyzed a number of existing approaches and chose the best one. Deep fusion generative adversarial networks (DF-GAN) is the network that performs better than its peers, at multiple levels, like the generated image quality or the respect of the giving descriptive text. Our model is trained on the CelebA dataset and text descriptions (generated by our algorithm using existing attributes in the dataset). The obtained results from our implementation show that the learned generative model makes excellent quantitative and visual performances, the model is capable of generating realistic and diverse samples for human faces and create a complete portrait with respect of given text description.