E-Mail Spam Filter Based on Unsupervised Neural Architectures and Thematic Categories: Design and Analysis

Author(s):  
Ylermi Cabrera-León ◽  
Patricio García Báez ◽  
Carmen Paz Suárez-Araujo
Keyword(s):  
Author(s):  
Olumide Babatope LONGE ◽  
Stella Chinye CHIEMEKE ◽  
Olufade F. Williams ONIFADE ◽  
Folake Adunni LONGE
Keyword(s):  

The inefficiencies of current spam filters against fraudulent (419) mails is not unrelated to the use by spammers of good-word attacks, topic drifts, parasitic spamming, wrong categorization and recategorization of electronic mails by e-mail clients and of course the fuzzy factors of greed and gullibility on the part of the recipients who responds to fraudulent spam mail offers. In this paper, we establish that mail token manipulations remain, above any other tactics, the most potent tool used by Nigerian scammers to fool statistical spam filters. While hoping that the uncovering of these manipulative evidences will prove useful in future antispam research, our findings also sensitize spam filter developers on the need to inculcate within their antispam architecture robust modules that can deal with the identified camouflages.


2014 ◽  
Vol 513-517 ◽  
pp. 2111-2114 ◽  
Author(s):  
Zong Jie Wang ◽  
Yi Liu ◽  
Zhong Jian Wang

The co-occurrence word emphasize the word and word internal relations, so its use can improve shortage from the hypothetical of Bayesian algorithm. To build Token Dictionary, Information Gain algorithm is used to choose Tokens, and Synonyms Dictionary is used to acquire more Tokens. By large amounts of training, the matching scores of Token are counted, according to the matching rate the Tokens that is valuable are selected, and the Token Dictionary is established. The proposed method is used to E-mail classification experiment, the results show that the accuracy of spam filter has a well improvement.


2015 ◽  
Vol 10 (1) ◽  
pp. 401-407 ◽  
Author(s):  
Seongwook Youn ◽  
Hyun-Chong Cho
Keyword(s):  

2015 ◽  
Vol 1 (2) ◽  
pp. 77
Author(s):  
Nurlina Nurlina ◽  
Irmayana Irmayana

Filter spam yang disediakan oleh situs penyedia layanan email seperti yahoo, gmail, aim mail, windows live hotmail, dan masih banyak lagi yang lainnya, merupakan fasilitas yang sangat bermanfaat bagi para usernya. Filter spam tidak akan berfungsi sebagaimana yang diharapkan oleh user client. Pada saat alamat email para client sudah pernah di subscribe dengan tujuan tertentu seperti misalnya untuk registrasi mailing list, newsgroup, dan lain sebagainya, maka alamat emailnya itu sudah tidak aman lagi dari para spammer. Pada dasarnya para admin mail server hanya menggunakan secara langsung filter spam yang disediakan oleh mail server yang diinstal, tanpa memberikan penyettingan tertentu yang dibutuhkan client sama sekali. Para user sendiri yang seharusnya lebih aktif dalam menyaring spam pada email yang digunakan dengan banyak cara. Penelitian ini memanfaatkan aplikasi mail client Thunderbird untuk menyaring spam dengan metode Bayesian sebagai kelanjutan dari menyaring spam yang sudah tersaring sebelumnya pada sisi mail server dan bertujuan menganalisis hasil pengklasifikasian email ham dan email spam pada mail server dan mail client. Disimpulkan bahwa nilai akurasi dan error filter spam pada mail server berhubungan dengan filter Spam-Assassin yang tidak disetting dan dikonfigurasi oleh adminnya menunjukkan hasil yang tidak memuaskan dibandingkan dengan filter spam metode bayesian pada mail client yang sudah di-training.Spam filters provided by your email service provider websites such as yahoo, gmail, AIM mail, windows live hotmail, and many others, is a very powerful feature for the user. The spam filter will not work as expected by the client user. at the time of the email address of the client has been ever subscribe to a specific purpose such as for registration, mailing lists, newsgroups, and so forth, then the email address is no longer safe from spammers. Basically the admin mail server directly using only the spam filter provided by the mail server is installed, without giving a specific setting it takes the client at all. The users themselves are supposed to be more active in the spam filter on the email that is used in many ways. This study utilizes Thunderbird mail client application to filter spam with Bayesian methods as a continuation of the spam filter that has been previously filtered on the mail server and to analyze the results of the classification of ham and spam e-mail on the mail server and mail client. It was concluded that the accuracy and error spam filter on the mail server associated with the filter Spam-Assassin is not be set and configured by the admin showed unsatisfactory results compared with the Bayesian method to filter spam mail client that is already in-training


Author(s):  
Nestor J. Zaluzec

The Information SuperHighway, Email, The Internet, FTP, BBS, Modems, : all buzz words which are becoming more and more routine in our daily life. Confusing terminology? Hopefully it won't be in a few minutes, all you need is to have a handle on a few basic concepts and terms and you will be on-line with the rest of the "telecommunication experts". These terms all refer to some type or aspect of tools associated with a range of computer-based communication software and hardware. They are in fact far less complex than the instruments we use on a day to day basis as microscopist's and microanalyst's. The key is for each of us to know what each is and how to make use of the wealth of information which they can make available to us for the asking. Basically all of these items relate to mechanisms and protocols by which we as scientists can easily exchange information rapidly and efficiently to colleagues in the office down the hall, or half-way around the world using computers and various communications media. The purpose of this tutorial/paper is to outline and demonstrate the basic ideas of some of the major information systems available to all of us today. For the sake of simplicity we will break this presentation down into two distinct (but as we shall see later connected) areas: telecommunications over conventional phone lines, and telecommunications by computer networks. Live tutorial/demonstrations of both procedures will be presented in the Computer Workshop/Software Exchange during the course of the meeting.


2005 ◽  
Vol 36 (1) ◽  
pp. 34
Author(s):  
SHARON WORCESTER
Keyword(s):  

2006 ◽  
Vol 37 (8) ◽  
pp. 56-57
Author(s):  
JOSEPH S. EASTERN
Keyword(s):  

Sign in / Sign up

Export Citation Format

Share Document