Probabilistic Dictionary Cracking Technology Based On Semantics

With the rapid development of the information technology, information security hasbeen paid more and more attention. As a branch of cryptography, password cracking is animportant research direction. Password cracking is essentially a study of the passworditself, which plays an important role in promoting the password-based securityauthentication system. Not only been used by the hackers to steal others’ information, butpassword cracking techniques also has many legitimate uses, such as application inelectronic evidence.Comparing to the fast development of other areas of computer science, the passwordcracking did not make many breakthroughs in recent years. Guessing attack is still themost popular method to crack. Basing on two important features of the password,human-memorable and statistical characteristic, we proposes a probabilistic dictionarycracking technology based on semantics in this paper, which is an improved dictionaryattack algorithm, in order to improve the success rate and efficiency of password cracking.The users selected passwords are somewhat predictable because of the human-memorable,which means when selecting a password, a user tend to choose string that is easy toremember for himself. Those strings are usually consist of elements which contain somesemantics. In addition, the password statistical property means that all kinds of passwordare not distributed randomly. There are some passwords used much more frequently thanthe other.To test the efficiency of the algorithm, a series of experiments are performed. Theresults show that, comparing to the pure brute force, the cracking rate increases from52%to64%, and comparing to John the Ripper, the most commonly used crack technology, itincreases from12%to45%.