Password dictionary generation method and device and computer equipment
A computer program and password technology, applied in the field of password security, can solve problems such as failure to meet user decryption needs, manpower, material resources and time, and reduce decryption success rate, to optimize the generation method, speed up decryption, and increase password coverage. and diverse effects
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0038] See figure 2 , The present invention provides a method for generating a password dictionary, including:
[0039] Step S110: Input the acquired real password sample set into the adversarial neural network model, where the real password sample set includes multiple real password samples.
[0040] Specifically, the adversarial neural network consists of two models: generator and discriminator. The generator learns real password samples and then generates fake password samples, and the discriminator also learns the identification of fake password samples through the real password samples. After repeated confrontation training, the generator produces more and more realistic password samples, and the discrimination accuracy of the discriminator is getting higher and higher, and the generator finally produces "real" passwords that are closer and closer to the real password samples. Among them, the set of real password samples can be obtained by searching for passwords leaked by t...
Embodiment 2
[0074] See Figure 4 As a preferred embodiment, the password dictionary generation method implemented by the present invention specifically includes the following steps:
[0075] Step S310: Collect a set of real password samples.
[0076] Step S320: Input the set of real password samples into the adversarial neural network, and initialize the adversarial neural network according to the training parameters.
[0077] Step S330: Start the adversarial neural network for adversarial training.
[0078] Step S340: Obtain fake password samples within the allowable similarity range generated when the adversarial neural network completes the training.
[0079] Step S350: Clean the fake password samples.
[0080] Step S360: According to the custom password combination rule, select a set of password samples that meet the rule and use it as a password dictionary.
[0081] It should be understood that although Figure 2-4 The steps in the flowchart are shown in sequence as indicated by the arrows, but...
Embodiment 3
[0083] See Figure 5 The present invention also provides a password dictionary generating device, which includes a real sample input module 410, an error acquisition module 420, and a password dictionary generating module 430.
[0084] The real sample input module 410 is used to input the acquired real password sample set into the adversarial neural network; wherein, the real password sample set includes multiple real password samples;
[0085] The error acquisition module 420 is used to acquire the return error obtained when the discriminator of the adversarial neural network discriminates the error between the real password sample and the current fake password sample generated by the generator;
[0086] The password dictionary generation module 430 is used to collect the fake password samples generated by the generator when the return error is less than the preset error as a collection to form a password dictionary.
[0087] The password sample dictionary generating device of the emb...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More