Biju Issac

Improved Bayesian Anti-Spam Filter Implementation and Analysis on Independent Spam Corpuses

Spam emails are causing major resource wastage by unnecessarily flooding the network links. Though many anti-spam solutions have been implemented, the Bayesian spam score approach looks quite promising. A proposal for spam detection algorithm is presented and its implementation using Java is discussed, along with its performance test results on two independent spam corpuses – Ling-spam and Enron-spam. We use the Bayesian calculation for single keyword sets and multiple keywords sets, along with its keyword contexts to improve the spam detection and thus to get good accuracy.

Secure unicast address resolution protocol (S-UARP) by extending DHCP

In IPv4 networks, address resolution protocol (ARP) is used when a source computer wants to find the MAC address corresponding to a particular IP address (of target computer), it wants to contact. The source broadcasts this request and all the hosts in the domain get that request. This broadcast helps the target to reply with the necessary details to the source. The IP-MAC mapping would then be stored in the ARP cache of the source, which remains valid for some time. This broadcasting is a decentralized way to get the IP address to MAC address mapping, which brings in consistent overhead to the network traffic. It is well documented that ARP process is also liable to ARP poisoning attacks. In this paper we seek to propose a secure unicast address resolution protocol (SUARP) which takes a centralized approach that reduces the broadcast overhead and makes ARP poisoning attacks difficult.

Detailed DoS attacks in wireless networks and countermeasures

IEEE 802.11 wireless networks have fuzzy boundaries, making it easy for an attacker to capture the transmitted packets. Also the attacker can send massive volume of illegitimate traffic and utilise system resources in a way that renders the system inoperable, thus denying access to authorised users. This paper is two fold in its description. Firstly, it describes some of the major vulnerabilities associated with wireless networks. Secondly, it demonstrates different methods of achieving Denial of Service (DoS) attacks in detail, as it applies to wireless networks and discusses and proposes different countermeasures to minimise the attacks.

Spam Detection Proposal in Regular and Text-based Image Emails

Spam, Emails are invading users without their consent and filling their mail boxes with Email trash. Priceless effort and time of the users and organizations are wasted in handling them. To circumvent anti-spam solutions, many spammers are sending spam Email with image-only content. In this paper we propose a spam detection approach in Emails with text and image contents. In the first part, a novel framework for extracting intelligent information from Emails with image content is presented and a prototype implementation is shown. In the second part, a proposal for multi-layered spam detection algorithm is presented, which enhances existing approaches

Appraising the corporate sustainability reports: text mining and multi-discriminatory analysis

The voluntary disclosure of the sustainability reports by the companies attracts wider stakeholder groups. Diversity in these reports poses challenge to the users of information and regulators. This study appraises the corporate sustainability reports as per GRI (Global Reporting Initiative) guidelines (the most widely accepted and used) across all industrial sectors. Text mining is adopted to carry out the initial analysis with a large sample size of 2650 reports. Statistical analyses were performed for further investigation. The results indicate that the disclosures made by the companies differ across the industrial sectors. Multivariate Discriminant Analysis (MDA) shows that the environmental variable is a greater significant contributing factor towards explanation of sustainability report.

Evaluation of Corporate Environmental Reports Using Data Mining Approach

Growing importance in corporate environmental reporting has undermined the metrics for its examination. Any such metrics are useful in measuring the effectiveness of reports. The main aim of the study in this paper is to examine and analyze the corporate environmental reports (N=539) using data mining approach. The assessment technique implemented in this study is in line with Global Reporting Initiative (GRI) disclosures. Our findings show that the data mining approach is a reliable and robust measurement. Further findings suggest that reports addressing the criteria differ from the nature of business. The study reveals the growing importance of addressing the various stakeholder groups, industrial base approach and organizational strategies in sustainable reporting.

The art of war driving and security threats - a Malaysian case study

The fact that wireless local area networks (WLAN) use radio spectrum for transmitting data has with it pros and cons. Mobility without wires and the ease to move around to connect to network resources has made IEEE802.11 WLANs quite popular. The users need to have a laptop with a wireless network adapter that negotiates with an access point. Once authenticated and associated with the access point, the user can easily move around in the transmission range of the access point without losing data or network connection enjoying bandwidths of the order of multiples of 10 Mega bytes. On the negative side, these wireless LANs tend to have fuzzy boundaries, making it easy for an intruder to capture these transmission signals with a receiving device fitted with a sensitive antenna. An analysis of these captured packets can be good news to the intruder. We try to investigate on war driving (an act of locating wireless networks from within a moving vehicle), on interception of transmission data from the located wireless LANs in some of the highways in our country and on doing a brief analysis of that, eventually discussing on the attacks and security precautions.

Analysis, Design and Implementation of a Web-Based Payroll Application Software

The need to develop customized application software is quite high now-a-days with the ever increasing automation process within many companies. In this paper, we analyze, design, implement and test a web-based payroll software for an external company with the Setup module, PI (personnel information) module, OT (over time) module, Payroll module and Report module. These modules and their implementation are discussed in detail. The software development life cycle followed is discussed to show how different phases are evolved and how they interact with each other to form the software that the client needs.

Pre-Authenticated and Adaptive IP Mobility for WLAN nodes

When a mobile node roams to another network/subnet, new resource reservation has to be made between the sender and receiver. If enough resources are not available, the quality of service may be disrupted and the connection may fail. If resources are reserved indiscriminately, then most of them would be unused and performance degradation would be the result. We would like to propose a pre-authenticated and adaptive micro-mobility management architecture in IP networks. We use pre-authentication to hasten the handoff process. Also the mobility predictor block used will help in reserving the resources aptly. The packets are generally classified based on the ToS field in IPv4 (or traffic class and the flow label fields in IPv6). Here we focus more on the special case of wireless local area networks (WLANs) installation within a building or in a restricted campus. In this paper, we propose a fast and adaptive IP micro mobility management scheme with mobility prediction for WLAN nodes. We also present some basic simulation on WLAN mobile node roaming