Method and system for recognizing fraud short message by using deep learning
What is Al technical title?
Al technical title is built by PatSnap Al team. It summarizes the technical point description of the patent document.
A deep learning and fraudulent SMS technology, applied in the field of communication security, can solve the problem of low recognition accuracy of fraudulent SMS algorithm
Inactive Publication Date: 2018-09-21
ZHEJIANG PONSHINE INFORMATION TECH CO LTD
View PDF5 Cites 21 Cited by
Summary
Abstract
Description
Claims
Application Information
AI Technical Summary
This helps you quickly interpret patents by identifying the three key elements:
Problems solved by technology
Method used
Benefits of technology
Problems solved by technology
[0006] The technical problem to be solved by the present invention is to provide a method and system for identifying fraudulent text messages using deep learning to solve the problem of low recognition accuracy caused by the ever-changing characteristics of fraudulent text messages
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more
Image
Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
Click on the blue label to locate the original text in one second.
Reading with bidirectional positioning of images and text.
Smart Image
Examples
Experimental program
Comparison scheme
Effect test
Embodiment 1
[0080] This embodiment provides a method for using deep learning to identify fraudulent text messages, such as figure 1 shown, including steps:
[0081] S11: Obtain the text data of the SMS sample and perform word segmentation processing;
[0082] S12: using Word2Vec to convert the text data after halving into word vectors;
[0083] S13: Using the LSTM algorithm to convert word vectors into sentence vectors;
[0084] S14: using the sentence vector as the input vector of the softmax classifier to train the deep learning model;
[0085] S15: Identify fraudulent text messages according to the output result of the softmax classifier of deep learning after training;
[0086] S16: If it is determined to be a fraudulent short message, intercept the short message.
[0087] In recent years, deep learning algorithms have been applied to the field of natural language processing and have achieved better results than traditional models. In natural language processing, the commonly use...
Embodiment 2
[0175] This embodiment provides a system that uses deep learning to classify fraudulent text messages, such as image 3 shown, including:
[0176] Processing module 21, is used for obtaining the text data of short message sample and carries out participle processing;
[0177] Word vector module 22, for adopting Word2Vec to convert the text data after the halving into word vector;
[0178] Sentence vector module 23, for adopting LSTM algorithm to convert word vector into sentence vector;
[0179] Training module 24, for using sentence vector as the input vector of softmax classifier to train deep learning model;
[0180] The identification module 25 is used to identify fraudulent text messages according to the output result of the trained deep learning softmax classifier.
[0181] This embodiment provides a system for effectively identifying fraudulent short messages by using a deep learning algorithm under massive data. When the SMS sender sends the SMS to the SMS network ...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more
PUM
Login to view more
Abstract
The invention discloses a method and a system for recognizing a fraud short message by using deep learning, used for solving the problem that an algorithm is not high in recognition accuracy due to the fact that the features of the fraud short message is ever-changing. The method comprises the steps of acquiring text data of a short message sample and performing word segmentation treatment; usingWord2Vec to convert the text data that is subjected to the word segmentation into word vectors; using an LSTM algorithm to convert the word vectors into sentence vectors; using the sentence vectors asinput vectors of a softmax classifier to train a deep learning model; and recognizing the fraud short message according to an output result of the trained and deeply learned softmax classifier. According to the method and system for recognizing the fraud short message by using deep learning provided by the invention, the capacity of accurately recognizing the fraud short message is improved.
Description
technical field [0001] The invention relates to the technical field of communication security, in particular to a method and system for identifying fraudulent short messages by using deep learning. Background technique [0002] SMS, as a carrier of information transmission among massive customers, sets up an effective channel for mutual communication. With the widespread use of short messages, the phenomenon of fraudulent information transmitted through short messages has become more and more serious, which has brought great inconvenience to mobile phone users for normal information exchange, and more and more users have been deceived and suffered unbearably. Word. [0003] While creating economic benefits, SMS has also brought serious social costs and cultural losses to operators. [0004] At present, the research on fraud SMS identification is mainly the combination of feature extraction and traditional machine learning algorithms, but feature extraction requires a lot o...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more
Application Information
Patent Timeline
Application Date:The date an application was filed.
Publication Date:The date a patent or application was officially published.
First Publication Date:The earliest publication date of a patent with the same application number.
Issue Date:Publication date of the patent grant document.
PCT Entry Date:The Entry date of PCT National Phase.
Estimated Expiry Date:The statutory expiry date of a patent right according to the Patent Law, and it is the longest term of protection that the patent right can achieve without the termination of the patent right due to other reasons(Term extension factor has been taken into account ).
Invalid Date:Actual expiry date is based on effective date or publication date of legal transaction data of invalid patent.
Login to view more 
Login to view more