Mar 12
3
Sidejacking Or Session Hijacking
Firesheep: A Mozilla Plug-in developed for Session Hijacking.
Vulnerabilities: Details of your login (User names, passwords) for Sites such as Facebook can be easily uncovered (i.e. your session details can be uncovered).
Root course: Although you login to certain sites which encrypt their log-in pages with SSL, it revert to HTTP as soon as the user leaves the log-in page, so that a hacker could easily hijack your session.
Solution: Only the use of full time TLS/SSL will keep a user safe from being Session Hijacked.
Business organizations should protect their customers and employees by means of
the following:
• Consider extending HTTPS to their whole website.
• Ensure all critical web applications use HTTPS.
• Use a trusted Certification Authority with a well recognized trust mark.
• Advise the customers via the website about the use of HTTPS and explain its importance.
• Enforce the use of VPNs for all critical web applications.
• Educate customers and colleagues about the dangers of Session Hijacking.
Enterprise Resource Planning System Overview
Why ERP?
- Improved economic conditions for companies.
- Improved liquidity.
- Better Financial and Business Reporting and there by better decision making.
- Reduced duplication of data and business processes.
- Improved interdepartmental and intradepartmental communication.
ERP – How they are expected to evolve?
A significant growth in ERP market was exhibited after the financial downtime of 2008. Firms requires higher ROI and Systems that are quicker and easier to setup and integrate.
It is vital important to see the impact of ERP on abidance with industry and government regulations, Mobile technologies and with newer Hi performance computing Platforms such as Clouds, will affect plans for ERP deployments in future.
What barriers exists for ERP?
- Budget constraints.
- Lack of Shareholder Interest.
- Lack of time available for implementation.
- Staff training issues.
- None
- Customization concerns.
- Database issues.
- Employee feeling of Resistance for change.
- Disaster recovery concerns.
- Others.
ERP, Detailed long term benefits
- Better financial reporting.
- Better Cash flow and liquidity leading to a reduction in customer debts.
- Better cash flow management across the organizations global infrastructure by enabling them to establish individual payment centers.
- Financial benefits could be realized in certain ways.
- Make savings in departmental budgets.
- Enabling to compare cost savings with revenue gains, so better decision making when setting targets.
- Enables to identify non-productive areas of business so that staff transfer can be done accordingly in order to improve non-productive areas..
- Enables to determine future trends based on past data so that sufficient inventories could be maintained accordingly.
- Risk analysis made easier. As a result backups and recovery techniques for an upcoming crisis could be pre-planned sophisticatedly.
- Enables to speed up monthly close process. As a result management information and analysis is made sooner so that any mitigation strategies can be taken in advance if deviations were detected with established baselines.
5. Improved Management Information Statistics
Information is integrated and available as a whole in a meaningful format rather than duplicated versions of true information. There for improved availability of information. This is a one single significant benefit.
Business Intelligence System is another key component of an ERP which help executives in their strategic plans as well as for daily operational business by in cooperating data searches, KPI dashboards, reporting and data mining. As a result, collaboration enabled among key decision makers by ERP allows resources to be synchronized with strategic plans which leads to significant impact on organization’s long term existence within its market with a higher market share and revenues.
A New route to ERP
Firms are willing to further upgrade their existing ERP system by adding further components such as CRM, HRM etc. And those that have not yet taken the challenge are starting to see the advantages of others who have taken the plunge.
In addition, firms are expressing their interest in extending their ERP systems to support mobile devices. The mobile market is continually leveraging itself with smart phones leading the way Mobile ERP is becoming a non-negotiable requirement in a distributed organization.
ERP on the Cloud
Despite its possible security and privacy risks, greater flexibility in deployment is one of the reasons that the SME sector is responding positively to Cloud ERP. The abrasive economic Climate of the last couple of years has made organizations to establish themselves as risk avoiders rather than risk seekers. They have to make sure that tightly restricted budgets of projects are not exceeded. Therefore risk and the cost of the traditional all-or-nothing ERP deployment is no longer an acceptable option.
The Cloud allows no longer having to worry about constant server updates and other computing issues. Organizations will be free to concentrate on innovation. Employees can access information wherever they are, rather than having to remain at their desks, ensuring more Mobility and Flexibility. Further it allows for unlimited scalability, in contrast to the traditional rigid, customer-owned private computer systems. Organizations are safeguarded from their full IT infrastructure because the Cloud’s inherent scalability integrates disaster recovery capabilities.
Social ERP ( Moving towards 360 Degree view of Customer data across the Enterprise)
By the term Social ERP seems almost silly. However it doesn’t seems really as it is. It is a concept of granting a value towards employees, customers, and other shareholders voices for organizations’ successful product or process quality and long term existence within its industry.
Social CRM is already established, allowing organizations to furnish social media conversations and relationships with customers. Social ERP is the next step in the integration of social media into business organizations, with the prospect to extend social customer data throughout the supply chain. Although there are raising concerns about privacy, it is becoming non-negotiable requirement when establishing long term customer relationship and loyalty.
Oct 11
30
Software Copy Protection
Andy Samberg did a hilarious f8 intro this morning, pretending to be Mark Zuckerberg. Such a fun moment.
The recent Facebook changes are game changing. For most this is scary; for us (and hopefully for you, too) it’s exciting! Facebook has managed to make our everyday activities even more social and has created a way for us to share our lives via a virtual scrapbook. This tool lets you feature all your old and new Facebook memories with integrated ways to express yourself through different “lifestyle apps.” !!!
Sep 11
15
Neural Network Input
1. If a data set shows that several/different inputs are mapping to same output, then we may use Artificial Neural Network Model, to model this issue.
[Note: If input and Output has one-to-one relationship we can model them using Rules. If input and output have one-to-many relationship you can use Fuzzy Logic ].
2. ANN input values must be expressed as very small quantities.
X = x1, x2, x3, x4, …, xn
For these purpose we divide all the inputs components by a big number like 100 or normalize the input vector.
3. Most of the image related data can be represented or prepared as an input for ANN by digitizing them in to 1 and 0 .
Image = { 0 1 0 0 0 0 1 0 0 1 1 0 0 0 0 0 }
4. When we have inputs with zeros (0) as a component, the The computation of Net value, ignore that component whatever the value of weights for that component.
Some inputs with all components zero ‘0’ may have non-zero output.
X = ( 0 0 0 ) =>> d (output) = ( 1 1 )
In this case, for whatever weights, you couldn’t achieve the desired output. There for we have the following solutions.
- To address this issue we can add extra non-zero component (e.g: -1 ), to each input vector. This unit is called a Bias.
As a matter of policy, Bias are added to all the neurons in the network.
Note: During Net calculations, neurons in the intermediary layers may also may generate their output as zero, which become the input for next layer. Therefore better to add Bias for all the Neurons.
- Use of Bi-polar activation function.
5. During a training session you never get the inputs repeatedly from the same class. Instead randomize over different classes. This help to develop a higher generalization.
If you train one class completely and train another class obviously error will increase at the transition point.
In order to randomize over classes, we must have a way to classify data.
Neural Network itself can be used to classify data.
In this case we can train the data set using un-supervised training by considering single layer Neural Network.
E.g. : – If we have a three (3) neuron in one layer during un-supervised training, we can identify which Input generate maximum output on which neuron. On this we have 3 different classes identified.
In this sense Neural Network training session, may be supported by another small neural network which work as the pre-processor for data classification.
.
.
.
Related Articles :
1. Choosing a propper Neural Network Architecture
2. Neural Network Training Issues
3. AForge.NET Framework for Artificial Neural Networks development
Aug 11
20
Terrain Analysis
Aug 11
6
Social Networking Challenges
Security : Organizations are concerned about the exposure of internal business systems to external entities.
Information that should be under tight control may be publicly exposed, either accidently or intentionally.
Control : Going forward, organizations need to decide what to share, how to share, and when to share.
The conventional wisdom has always been that controlling information is better than sharing it.
Lack of integration of Social software with other tools used by employees.
The cost of navigation back and forth between multiple applications and separate windows can significantly reduce the organizational performance in long run.
Trust and Privacy : Concerns and unsease with new methods for interacting with (unknown) contacts.
Jul 11
5
High Performance Computing
” It is not about ‘harder, better, faster, stronger “, building more and more powerful computers, It’s about linking and using the resources optimally you have. Or is it?
The most commonly available High Performance Computing (HPC) facilities of today are the descendants of the super computers in the past. They have many processors, and complicated architectures, and process data in parallel. Now a days the fastest computer on the planet isn’t a computer that we recognize it at all. It should be a loose cloud of machines spread across a network /internet engaged in peta FLOPS of (Floating point operations per second) speed.
At a glance
Multiprocessor systems – Most computers now come with more than one processor.
Cluster Computing – A set of linked computers networked as a single entity. Runs the same operating systems, linked on a local area network, and often in the same room. Can be used to maintain reliability, robustness. E.g. Load balancing clusters.
Grid computing – Cluster computing with loosely coupled computer nodes. Often in different places. The compuers can have varied software installed on various types of hardwrae platforms, and user depends on middleware to determine what part of the grid run what code, by further making use of spare cycles on machines which do other daya to day activities.
Cloud computing – A Distributed Computing paradigm. Often relay on grid computing systems, but more loosely coupled and doesn’t necessarily involve central administration. For example Google docs, runs on a grid somewhere,but you can access your files and aplications from any internet connection.
Today a system to be considered as a supercomputer it speed has to be approaching teraFLOPS. In order to determine how fast a particular High Performance Computing setup will run is not like going in to PC world and choosing the fastest machine you can. Different problems can be more suited to different architectures, and a certain level of optimization is required. By running Linpack benchmark for different problem sizes we can detarmine a real maximum performance of a specific High Performance Computing system for a certain type of problem.
The Future
- Green Computing – It is critical to getting the most computing power output of your system but with the smallest impact on the environment. For HPC practitioners this is a key issue, as generally a room of servers with a high workload can produce serious amount of heat. Therefore Supercomputers can no longer focus only on raw performance, to be commercialized it should be energy efficient. The next generation of Supercomputers are Hybrid systems that combine different processor architectures for better performance and energy efficiency.
- Hybrid Systems – Even low coupling, high cohesion HPC systems have power implications, the costs are passed on to the volunteers devoting their spare compute cycles to a project. There for, this solution server as a remedy for the power implication issues in-cooperated with HPC platforms.
- Semantic Web/Networks .
- Parallel Processing.
1. What-If analysis feature in Exacative dashboards
Change the sales/expenses graph values by dragging and dropping the data points up and down. The corresponding Profit value will be shown in the Profit graph in the next dimension.
2. Drill-Down effect
Select a set of data points by pressing the mouse point at one position and pull it back until a rectangular region with a set of data points to be selected. The in depth description of selected data will be shown in the other chart/dimensition.
Jun 11
29
Revolutionary Fuzzy Logic
” The concept of Fuzzy Logic (FL) was conceived by Lotfi Zadeh, a professor at the University of California at Berkley, and presented not as a control methodology, but as a way of processing data by allowing partial set membership rather than crisp set membership or non-membership. This approach to set theory was not applied to control systems until the 70′s due to insufficient small-computer capability prior to that time. Professor Zadeh reasoned that people do not require precise, numerical information input, and yet they are capable of highly adaptive control. If feedback controllers could be programmed to accept noisy, imprecise input, they would be much more effective and perhaps easier to implement.
In this context, FL is a problem-solving control system methodology that lends itself to implementation in systems ranging from simple, small, embedded micro-controllers to large, networked, multi-channel PC or workstation-based data acquisition and control systems. It can be implemented in hardware, software, or a combination of both. FL provides a simple way to arrive at a definite conclusion based upon vague, ambiguous, imprecise, noisy, or missing input information. FL’s approach to control problems mimics how a person would make decisions, only much faster.” by Encoder.
Applications
1. Fuzzy logic Anti-Sway Controller for Container Crane Control
2. Fuzzy logic for Automated Destroy Missions
Jun 11
29
Network Programming – An overview
“In computing, network programming, essentially identical to socket programming or client–server programming, involves writing computer programs that communicate with other programs across a computer network. The program or process initiating the communication is called a client process, and the program waiting for the communication to be initiated is the server process. The client and server processes together form a distributed system. The communication between the client and server process may either be connection-oriented (such as an established TCP virtual circuit orsession), or connectionless (based on UDP datagrams).” Wikipedia.
Sample Application
A Multi-Threaded HTTP Server Using Sockets in Java
Jun 11
29
www – Tips, Technologies
Tips & Latest Technologies
1. How to design your web site in Dreamweaver ?
2. Rich Internet Applications (RIAs)
3. Why Abobe being popularized in RIA community?
Jun 11
23
Geographic Information Systems
A geographic information system (GIS), geographical information system, or geospatial information system is a system designed to capture, store, manipulate, analyze, manage and present all types of geographically referenced data. Today, many fields applies GIS technology to analyze complex situations and create solutions across disciplines.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
GIS gives you the capabilities to improve the understanding of the situation, and provides the intelligence to support your need for mission success. GIS helps you increase efficiency, reduce costs, improve coordination, and deliver quality.
A Geographic Information System for Epidemiology
1. How GIS can benifit | Epedimiology
2. Data Requirements | Data Sources
3. Factors for effective GIS Solutions
Artificial Neural Network (ANN) is a model of a human brain. According to Biology brain is a device with a millions of Neurones that are connected to form a Network.
Neurons receive inputs and do the processing inside it. and pass the output to connected neurons through the connections. Neurons work as processors and connections work as memories to locate as learning effect.
The neurons in the brain (processors or CPU) work parallel to manipulate signals such a s sounds, visuals, smells, body cells. Perhaps the best example of most complex parallel computing device is the Brain.
Brain main features
1. Brain will fed up.
2. Abstraction – Picking some specific features.
3. Classification – Putting in to different classes.
4. Generalization – Represent large number of things by small number of things.
5. Learning by training.
Due to these reasons Neural Networks can be used in the real world. All Neural Network applications must be given a large number of data. Else we cannot achieve a better Generalization.
The kind of data could be with lot of noise, incompleteness, non-algorithmic nature . (e.g. Signature, Medical Images, Photos etc.).
It can indentify previously unseen patterns, trends, relationships through the abstraction property. (e.g. Data Mining).
Classification – Putting data in to various classes. E.g. – Object Recognition, Market Segmentation.
Special Articles
1. AForge.NET Framework for Artificial Neural Networks development
2. Choosing a propper Neural Network Architecture
3. Neural Network Training Issues
Subscribe to our Tech E-News letter and be aware with the Latest Technologies