Online Spamming (Seham ALHAZMI)

Introduction:
The modern scenario of the business is enriched with the idea of online activities. On one hand these online practices have made the conventional practices of conducting business easily and on other there are few very serious issues associated with the problem as well. In the similar capacity the spamming ss the new issue that is faced by the usual customer from the online marketing threads (Xavier Carreras, Lluís Màrquez, 2001). At many times these threads spread the spamming activity and at time they are real. The following post will discuss about different areas of spamming under the name of marketing for business. The post will also illustrates the associated issues and platforms for the spammers to create these issue for a general internet user.

Spam Classification for Online Discussion:
Back to the old times, when we have watched a drama or a sport game we would prefer to talk to our neighbours and friends as soon as possible. We believe that we could share our happiness with others in the talks and discussions. When we need some helps such as clogged pips or babysitting, those neighbours would be our first choice before other options run through the head.
However, those have all been changed by a great invention “computer”. None of us is able to escape from this life style revolution caused by the birth of computer and everyone is involved in this significant change due to the Internet.

Problem Declaration:
Have you been wondering where are those sales men going? They used to knock our doors and start with a very polite asking “Excuse me, can I have you a few seconds?” Those “a few seconds” will soon be increased to a couple of minutes. If you don‟t stop them and throw them out of your house immediately, you will be out of your mind by their endless stories (Richard O. Duda, Peter E. Hart, David G 2000). Even if they don‟t visit you in the front door, they could also “concern” your needs by calling you. This situation is even worse that it is unpreventable before you pick up the phone. However, why don‟t we have sales calls and door-to-door sales men anymore? Are they still in the same time and space where we stay? Yes, they do. They just changed the way to make business. We shall solve the mystery in the following parts.
Living in the 21th century, we no longer enjoy those great afternoons with friends and families by drinking a cup of coffee or tees. After a tough day, we are more likely to join in those conversations and discussions what we concern through Internet. There we have a lot of virtual communities which promise to provide nice and safe places for us to give opinions.
Without face-to-face conversation, participants may be in the different age groups, social classes or different places. They have chances to talk to whom they hardly have chance to meet during the work or in their normal life or even in their spaces. In this case, the platform gives every participant an equal opportunity to present ideas and thoughts. It also gives opportunities to someone who does not intent to participate in discussions.

Growing Problems with Growing Trends:
As the number of Cyber Citizens is growing rapidly in these recent years, Internet brought us much more surprises than we used to expect, not only explosive information but also explosive problems. On one hand, people enjoy the good gossip with some John Doe from any part of the world; on the other hand, sales men are trying their best to interrupt into your conversations to get as much attention as possible.

Spammer or an Online Saleman?
Now, we may give this kind of sales men a new name “spammer”. Once we all clear about the purpose of spamming and how they will spam, we could start to think about the solution to filter spam from those platforms.

Example Analysis:
Consider these two situations: 1. sales men visit every house in the community taking a couple of weeks; 2. sales men send messages which contain every detail of the product to a discussion which will be browsed by dozens of people in a minute ( E. M. Voorhees, 2002) . Which one is better in your opinion? Do you thing in our opinion, there is no doubt that the second one is way better than the first one which not only saves the time and energy for the sales men but also gives guarantee to products from being damaged during the demonstration travelling.
From the above discussion we could understand that why sales men pick Internet to broadcast their advertisements. There is no fault to take a channel as low-cost as possible to do business in the first place. However, it is unacceptable that if the business has interrupted into our normal life.
Those sales men not only broadcast their advertisements they actually push people to read them and this is the reason we call them ― spam. Then we shall know how these spammers spam our Internet.

Here I show an example in the above Figure 1-2, with the “help” of a spammer, in which we may have a clear impression about in what way and where a spammer will post messages. The discussion is a fragment of large discussion happened on the wall page of a Facebook group user. All those posts are about the topic referring to “Pepsi” excluding the second reply which had only a link address. Apparently, the replier wanted participants to believe that the web site is something related to the topic. However, we found out that this web site had nothing to do with Pepsi after we tested it. Therefore, we consider this post is a spam.
The last one is the most challenging situation which we will encounter in this essay. In this situation, the spammer post messages which are relevant to the topic started by the first person, but the purpose of the posting is to publicizing a certain service or product of their own.

Platform for Spamming:
There are several places that we can find spam in the Internet such as E-mails, news groups, forums and etc. If you want to broadcast advertisements to as many people as possible, you need to choose the most crowd places and Social Media Conversation is one of those places (S. B. Kotsiantis, 2007).
In the past few decades, researchers have already put lots of strength on E-mail abuse and successfully developed kinds of methods to prevent the E-mail spam distribution. For instance, researchers in proposed a method using boosting tree to filter spam Email where AdaBoost gives quite good result for spam filtering (Niu, Y., Wang, Y. M., Chen, H., Ma, M., and Hsu, F, 2007). And researchers in compared Naïve Bayes classification to Memory-based classification for spam filtering with the conclusion that both approaches gain very high classification accuracy and memory-based approach appears to be more viable by careful configuration.

Social Media and Spamming:
Recently, as the Internet stepping into hundreds of thousands of ordinary people‟s life, the abuse of social media to advertise or spread bulk information attracts researchers‟ attention. Researchers from UMBC are doing great work in Social Media spam investigations, the spam messages are growing all the time through all new kinds of Social Media channels (planet.socialmediaresearch, 2012).
In 2007, scientists have already tried to filter spam messages from forums in in which presented us a quite good context-based approach to detect spam.
That even some researchers have already started to analyze Social Media spam, its still a quite new topic and field to be explored.

References:
[1] Xavier Carreras, Lluís Màrquez, Boosting trees for clause splitting, Proceedings of the 2001 workshop on Computational Natural Language Learning, p.1-3, July 06-07, 2001, Toulouse, France.
[2] About planet.socialmediaresearch.org. [Online]. Available: http://planet.socialmediaresearch.org.
[3] Niu, Y., Wang, Y. M., Chen, H., Ma, M., and Hsu, F. A Quantitative Study of Forum Spamming Using Context-based Analysis. In Proc. Network and Distributed System Security (NDSS) Symposium, February 2007.
[4] Richard O. Duda, Peter E. Hart, David G. Stork, Pattern Classification (2nd Edition), WileyInterscience, 2000.
[5] S. B. Kotsiantis, Supervised machine learning: A review of classification techniques, Overview paper, July 2007.
[6] E. M. Voorhees. Overview of the TREC 2002 question answering track. In Proceedings of the Eleventh Text REtrieval Conference (TREC 2002)2002.
[7] Ian Soboroff and Donna Harman, Novelty detection: the TREC experience, Proceedings of the conference on Human Language Technology and Empirical Methods in Natural Language Processing, p.105-112, October 06-08, 2005, Vancouver, British Columbia, Canada.

Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License