Dark web content analysis and identification
In some examples, dark web content analysis and identification may include ascertaining data that includes text and images, and analyzing the data by performing deep learning based text and image processing to extract text embedded in the images, and deep embedded clustering to generate clusters. Clusters that are to be monitored may be ascertained from the generated clusters. A determination may be made as to whether the ascertained data is sufficient for classification. If so, a deep convolutional generative adversarial networks (DCGAN) based detector may be utilized to analyze further data with respect to the ascertained clusters, and alternatively, a convolutional neural network (CNN) based detector may be utilized to analyze the further data with respect to the ascertained clusters. Based on the analysis of the further data, an operation associated with a website related to the further data may be controlled.
1. A dark web content analysis and identification apparatus comprising:
a data receiver, executed by at least one hardware processor, to
ascertain data that includes text and images;
a deep learning based data analyzer, executed by the at least one hardware processor, to
analyze the ascertained data by performing
deep learning based text and image processing on the ascertained data to extract text embedded in the images, and
deep embedded clustering with respect to the ascertained text, the images, and the text extracted from the images to generate a plurality of clusters;
a data enricher, executed by the at least one hardware processor, to
ascertain clusters, from the plurality of generated clusters, that are to be monitored;
an intelligence applicator, executed by the at least one hardware processor, to
ascertain further data that is to be analyzed,
determine whether the ascertained data is sufficient for classification,
based on a determination that the ascertained data is not sufficient for classification, utilize a deep convolutional generative adversarial networks (DCGAN) based detector to analyze the further data with respect to the ascertained clusters, and
based on a determination that the ascertained data is sufficient for classification, utilize a convolutional neural network (CNN) based detector to analyze the further data with respect to the ascertained clusters; and
an insights based controller, executed by the at least one hardware processor, to
control, based on the analysis of the further data, an operation associated with a website related to the further data.
2. The dark web content analysis and identification apparatus according to claim 1 , wherein the deep learning based data analyzer is executed by the at least one hardware processor to analyze the ascertained data by performing deep embedded clustering with respect to the ascertained text, the images, and the text extracted from the images to generate the plurality of clusters by:
analyzing, for the ascertained text and the text extracted from the images, combine continuous bag of words (CBOW) based similarity;
analyzing, for the ascertained images, convolutional neural network (CNN) based similarity; and
generating, based on the CBOW based similarity and the CNN based similarity, the plurality of clusters.
3. The dark web content analysis and identification apparatus according to claim 1 , wherein, based on the determination that the ascertained data is not sufficient for classification, the intelligence applicator is executed by the at least one hardware processor to utilize the DCGAN based detector to analyze the further data with respect to the ascertained clusters by:
utilizing the DCGAN based detector to analyze the further data with respect to the ascertained clusters to identify a similar source with respect to the further data.
4. The dark web content analysis and identification apparatus according to claim 3 , wherein the intelligence applicator is executed by the at least one hardware processor to utilize the DCGAN based detector to analyze the further data with respect to the ascertained clusters to identify the similar source with respect to the further data by:
utilizing the DCGAN based detector to analyze the further data with respect to the ascertained clusters to identify the similar source that includes a similar website with respect to the further data.
5. The dark web content analysis and identification apparatus according to claim 4 , wherein the intelligence applicator is executed by the at least one hardware processor to utilize the DCGAN based detector to analyze the further data with respect to the ascertained clusters to identify the similar source that includes the similar website with respect to the further data by:
determining the similarity of the similar website with respect to the further data based on at least one of a similarity of images included in the similar website, or a similarity of an organization of the similar website.
6. The dark web content analysis and identification apparatus according to claim 1 , wherein, based on the determination that the ascertained data is sufficient for classification, the intelligence applicator is executed by the at least one hardware processor to utilize the CNN based detector to analyze the further data with respect to the ascertained clusters by:
utilizing the CNN based detector to analyze the further data to score the further data.
7. The dark web content analysis and identification apparatus according to claim 1 , wherein the insights based controller is executed by the at least one hardware processor to control, based on the analysis of the further data, the operation associated with the website related to the further data by:
blocking access to the website related to the further data.
8. The dark web content analysis and identification apparatus according to claim 1 , wherein the insights based controller is executed by the at least one hardware processor to control, based on the analysis of the further data, the operation associated with the website related to the further data by:
generating an alert with respect to the website related to the further data.
9. The dark web content analysis and identification apparatus according to claim 1 , wherein the insights based controller is executed by the at least one hardware processor to control, based on the analysis of the further data, the operation associated with the website related to the further data by:
generating, based on at least one of a type or a severity of a corresponding cluster of the ascertained clusters, an alert with respect to the website related to the further data.
10. The dark web content analysis and identification apparatus according to claim 1 , wherein the intelligence applicator is executed by the at least one hardware processor to:
train, based on the plurality of generated clusters, the DCGAN based detector to analyze the further data with respect to the ascertained clusters.
11. The dark web content analysis and identification apparatus according to claim 1 , wherein the intelligence applicator is executed by the at least one hardware processor to:
train, based on the plurality of generated clusters, the CNN based detector to analyze the further data with respect to the ascertained clusters.
12. A method for dark web content analysis and identification comprising:
ascertaining, by at least one hardware processor, data that includes text and images;
analyzing, by the at least one hardware processor, the ascertained data by performing
deep learning based text and image processing on the ascertained data to extract text embedded in the images, and
deep embedded clustering with respect to the ascertained text, the images, and the text extracted from the images to generate a plurality of clusters;
training, by the at least one hardware processor and based on the plurality of generated clusters, a deep convolutional generative adversarial networks (DCGAN) based detector and a convolutional neural network (CNN) based detector;
ascertaining, by the at least one hardware processor, clusters, from the plurality of generated clusters, that are to be monitored;
ascertaining, by the at least one hardware processor, further data that is to be analyzed;
determining, by the at least one hardware processor, whether the ascertained data is sufficient for classification;
based on a determination that the ascertained data is not sufficient for classification, utilizing, by the at least one hardware processor, the DCGAN based detector to analyze the further data with respect to the ascertained clusters;
based on a determination that the ascertained data is sufficient for classification, utilizing, by the at least one hardware processor, the CNN based detector to analyze the further data with respect to the ascertained clusters; and
controlling, by the at least one hardware processor and based on the analysis of the further data, an operation associated with a website related to the further data.
13. The method according to claim 12 , wherein analyzing, by the at least one hardware processor, the ascertained data by performing deep embedded clustering with respect to the ascertained text, the images, and the text extracted from the images to generate the plurality of clusters further comprises:
analyzing, for the ascertained text and the text extracted from the images, combine continuous bag of words (CBOW) based similarity;
analyzing, for the ascertained images, CNN based similarity; and
generating, based on the CBOW based similarity and the CNN based similarity, the plurality of clusters.
14. The method according to claim 12 , wherein, based on the determination that the ascertained data is not sufficient for classification, utilizing, by the at least one hardware processor, the DCGAN based detector to analyze the further data with respect to the ascertained clusters further comprises:
utilizing the DCGAN based detector to analyze the further data with respect to the ascertained clusters to identify a similar source with respect to the further data.
15. The method according to claim 14 , wherein utilizing the DCGAN based detector to analyze the further data with respect to the ascertained clusters to identify the similar source with respect to the further data further comprises:
utilizing the DCGAN based detector to analyze the further data with respect to the ascertained clusters to identify the similar source that includes a similar website with respect to the further data.
16. The method according to claim 15 , wherein utilizing the DCGAN based detector to analyze the further data with respect to the ascertained clusters to identify the similar source that includes the similar website with respect to the further data further comprises:
determining the similarity of the similar website with respect to the further data based on at least one of a similarity of images included in the similar website, or a similarity of an organization of the similar website.
17. A non-transitory computer readable medium having stored thereon machine readable instructions, the machine readable instructions, when executed by at least one hardware processor, cause the at least one hardware processor to:
ascertain data that includes text and images;
analyze the ascertained data by performing
deep learning based text and image processing on the ascertained data to extract text embedded in the images, and
deep embedded clustering with respect to the ascertained text, the images, and the text extracted from the images to generate a plurality of clusters;
ascertain clusters, from the plurality of generated clusters, that are to be monitored;
ascertain further data that is to be analyzed;
determine whether the ascertained data is sufficient for classification;
based on a determination that the ascertained data is not sufficient for classification, utilize a deep convolutional generative adversarial networks (DCGAN) based detector to analyze the further data with respect to the ascertained clusters;
based on a determination that the ascertained data is sufficient for classification, utilize a convolutional neural network (CNN) based detector to analyze the further data with respect to the ascertained clusters by utilizing the CNN based detector to analyze the further data to score the further data; and
control, based on the analysis of the further data, an operation associated with a website related to the further data.
18. The non-transitory computer readable medium according to claim 17 , wherein the machine readable instructions to control, based on the analysis of the further data, the operation associated with the website related to the further data, when executed by the at least one hardware processor, further cause the at least one hardware processor to:
block access to the website related to the further data.
19. The non-transitory computer readable medium according to claim 17 , wherein the machine readable instructions to control, based on the analysis of the further data, the operation associated with the website related to the further data, when executed by the at least one hardware processor, further cause the at least one hardware processor to:
generate an alert with respect to the website related to the further data.
20. The non-transitory computer readable medium according to claim 17 , wherein the machine readable instructions to control, based on the analysis of the further data, the operation associated with the website related to the further data, when executed by the at least one hardware processor, further cause the at least one hardware processor to:
generate, based on at least one of a type or a severity of a corresponding cluster of the ascertained clusters, an alert with respect to the website related to the further data.