Customer retention has become a necessity in many markets, including mobile telecommunications. As it becomes easier for customers to switch providers, the providers seek to improve prediction models in an effort to intervene with potential churners. Many studies have evaluated different models seeking any improvement to prediction accuracy. This study proposes that the attributes, not the model, need to be reconsidered. By representing call detail records as a social network of customers, network attributes can be extracted for use in various traditional prediction models. The use of network attributes exhibits a significant increase in the area under the receiver operating curve (AUC) when compared to using just individual customer attributes.

Mining telecommunication networks to enhance customer lifetime predictions

Nephrotic syndrome (NS) is one of the most common causes of chronic kidney disease in children. Mostly, NS is controlled by steroids. In spite of this, 10% to 20% of the patients have steroid resistant NS (SRNS), and the rest have steroid sensitive NS (SSNS). Eighty-seven children with NS (66 SSNS; 21SRNS) were retrospectively studied within the past 10 years in Taif region of Saudi Arabia. Regarding outcome, 66 (76%) patient responded to the steroid therapy, while 21 (24%) patients characterize as SRNS. Out of 66 SSNS, 25 children revealed complete remission, 26 were diagnosed as steroid dependent, and 15 were identified as frequent relapses. Among 21 SRNS biopsied patients, focal segmental glomerulosclerosis was the most common histological patterns of NS followed by the membranoproliferative glomerulonephritis pattern, which is significantly different from other studies performed in the Kingdom of Saudi Arabia. Out of 21 SRNS patients, 13 developed chronic kidney disease, 4 got end-stage renal disease, and 4 patients died.

Patterns of Childhood Steroid-Sensitive and Steroid-Resistant Nephrotic Syndrome in Saudi Children.

—In the telecommunication industry, being able to predict customers’ behavioral pattern to successfully design and recommend a suitable tariff plan is the ultimate target. The behavioral pattern has a vital connection with the customers’ demographic background. Different researches have been done based on hypothesis testing, regression analysis, and conjoint analysis to determine the interdependencies among them and the effects on the customers’ behavioral needs. This has presented us with ample scope for research using numerous classification-based techniques. This work proposes a model to predict customer’s behavioral pattern by using their demographic data. This model was built after investigating various types of classification-based machine learning techniques including the traditional ones like decision tree, k-nearest neighbor, logistic regression, and artificial neural networks along with some ensemble techniques such as random forest, adaboost, gradient boosting machine, extreme gradient boosting, bagging, and stacking. They are applied to a dataset collected using a questionnaire in India. Among the traditional classifiers, decision tree gave the best result of 81% accuracy and random forest showed the best result among the ensemble learning techniques with an accuracy of 83%. The proposed model has shown a very positive outcome in predicting the customers’ behavioral pattern.

A Machine Learning Model for Personalized Tariff Plan based on Customer’s Behavior in the Telecom Industry

Cassandra is a NoSQL database having a peer-to-peer, ring-type architecture. Cassandra offers fault-tolerance, data replication for higher availability as well as ensures no single point of failure. Given that Cassandra is a NoSQL database, it is evident that it lacks research that has gone into comparatively older and more widely and broadly used SQL databases. Cassandra’s growing popularity in recent times gives rise to the need to address any security-related or recovery-related concerns associated with its usage. This review paper discusses Cassandra’s existing deletion mechanism and presents some identified issues related to backup and recovery in the Cassandra database. Further, failure detection and handling of failures such as node failure or data center failure have been explored in the paper. In addition, several possible solutions to address backup and recovery, including recovery in case of disasters, have been reviewed.

/pdf/backup-and-recovery-mechanisms-of-cassandra-database-a-16ctbrplxa.pdf

Backup and Recovery Mechanisms of Cassandra Database: A Review

In today's fierce telecommunications market competition, customer chum is very severe. In order to retain customers, telecommunications companies have made various attempts from various data and consumption characteristics analysis to big data analysis. However, since the actual situation of customer churn is very complicated, how to predict customer churn accurately and quickly is a difficult problem. After the researchers successfully conducted big data analysis of customer churn and successfully retained customers, in this article, the researchers mainly compared several commonly used algorithms in order to find a better algorithm for big data analysis of telecommunications customer churn. Compare and analyze the accuracy and efficiency of these several algorithms and suggest that the business support staffs of telecommunications companies adopt major methods for big data analysis. The researcher found that the Decision Tree (CART) algorithm is better for the prediction of customer churn and guided other branch staffs to predict customer churn and retain customers in a timely manner. This kind of big data analysis can be used to retain customers in the telecommunications industry.

https://iopscience.iop.org/article/10.1088/1757-899X/1077/1/012045/pdf

Comparison of Main Algorithms in Big Data Analysis of Telecom Customer Retention

Customer retention in Telecom market is a big research challenge in developed as well as developing economies as the market is almost saturated as well as competitive with large number of local and global service providers. It is also well known that from business point of view retaining an existing customer is much less costly than acquiring a new one. Hence retaining existing customer by making them loyal is crucial for the sake of existence of the telecom providers. Customers can be made loyal by making them satisfied customers. One of the important factors that make a customer satisfied is the value for money as well as proper prepaid recharge plans that will suit their requirements. An association based rule mining model is proposed in this research work to enable telecom operators to choose proper recharge combo plans that will be mostly appropriate for the customers. The main objective of this analysis is to identify the frequent recharge-types (itemset) based on the support and confidence value from recharge history dataset. This will allow telecom providers to focus on creating combo plans by aggregating the mostly used service-categories.

Designing combo recharge plans for telecom subscribers using itemset mining technique

Education involves perception of colour and unfortunately some of us are not blessed with this gift of proper vision. Lacking the ability to distinguish certain colours is commonly known as Colour Blindness. In schools and colleges, colour blind students are not supported much in the classroom, some lose confidence and struggle to cope. Often colour blind teachers are unable to distinguish important colours of pens or markers and for an efficient and interactive learning environment it is seen that the results are tilted in favour of classrooms where multiple colours are used on the board. The numbers corresponding to people having this disease(or its variations) are compelling enough to make it highly imperative for us to seek a solution for it. Hence, we propose a software which can be used as a virtual teaching and virtual painting platform by these differently abled people. Built on the Graphical User Interface Development Environment(GUIDE) in the numerical computing environment MATLAB, this system helps the user to identify the colour of the object (marker) withwhich he can teach virtually by moving it in air.These movementsget traced on the screen which serves as a virtual white board. I. Proposed System The software we propose can be used as a dual purpose tool. 1. It can be used as a virtual painting tool for the color blind children 2. It can be used as a virtual teaching tool for the color blind teachers. Both purposes are served by a set of functionalities which use the basic web camera of a computer system only as a resource.  The system using the web camera records the video of the user who holds an object (usually a marker) of a particular color( one of Red, Green, Blue, Cyan, Magenta and Yellow) as a tool used for drawing or writing on the Virtual White Board present on the interface.  This video is displayed separately on the left side of the interface for the user’s convenience in case he/she wants to track his/her hand movements. The user holds this object in front of the camera for some time for the system to detect the color of the object on a real time basis.  As soon as the color is detected, the user is prompted about its information on the virtual white board mentioned above. After this, the paint tool gets activated which traces the movement of the object in the user’s hands and paints the area (pixels) on the white board according to it.  The tracing of movement is visible to the user and he/she is able to see a rectangular bounding box, surrounding the colored portion of the object, which moves along with the object simultaneously depending on the user’s hand movement.  The painting occurs with the color detected by the system in the initial phase. In case the user wants to add a space or a pause in the process of writing or drawing then he/she can click on aPause button present at the bottom of the interface which halts the paint tool accordingly.  The user can also clear the virtual white board using a Clear button.  While writing or drawing if the user feels the need to change the color which he wants to use then he/she can do that using a Change Color button present on the interface. Once this button is clicked, the system re runs the detection tool to detect the color of the next object to be used by the user after which the user can continue his work.  The user can also save his work using a Save button present on the interface which saves the image of the virtual white board in .jpeg format.  The system also allows the user to paint or write on an image or a set of images which can be used as the same virtual board instead of the plain white one given before. This can be done by clicking on a button named Open present on the same interface. It also allows us to open a set of multiple images . Virtual Teaching and Painting Platform for the Colour Blind www.iosrjournals.org 2 | Page  Open functionality provides an experience equivalent to that of a power-point presentation on which the user can mark or write anything virtually. A set of buttons Next and Previous allow the user to switch between the set of images selected by the user before.  After the user has finished working with the system he/she can close/terminate it using a button named Close provided for the same on the interface. The juxtaposition of all the aforementioned functionalities in our proposed would serve as a handy and uniqueteaching or drawing tool which can prove to be highly effective in addressing the needs of the color blind children as well as the teachers. II. Why Matlab? MATLAB (matrix laboratory) is a multi-paradigm numerical computing environment and fourthgeneration programming language. Developed by MathWorks, MATLAB allows matrix manipulations, plotting of functions and data, implementation of algorithms, creation of user interfaces, and interfacing with programs written in other languages, including C, C++, Java, and Fortran. Although MATLAB is intended primarily for numerical computing, an optional toolbox uses the MuPAD symbolic engine, allowing access to symbolic computing capabilities. An additional package, Simulink, adds graphical multi-domain simulation and Model-Based Design for dynamic and embedded systems. MATLAB ® is a high-level language and interactive environment for numerical computation, visualization, and programming. Using MATLAB, you can analyze data, develop algorithms, and create models and applications. The language, tools, and built-in math functions enable you to explore multiple approaches and reach a solution faster than with spreadsheets or traditional programming languages, such as C/C++ or Java TM . Key Features:  High-level language for numerical computation, visualization, and application development  Interactive environment for iterative exploration, design, and problem solving  Mathematical functions for linear algebra, statistics, Fourier analysis, filtering, optimization, numerical integration, and solving ordinary differential equations  Built-in graphics for visualizing data and tools for creating custom plots  Development tools for improving code quality and maintainability and maximizing performance  Tools for building applications with custom graphical interfaces  Functions for integrating MATLAB based algorithms with external applications and languages such as C, Java, .NET, and Microsoft ® Excel ® III. Advantages Of Matlab Over Open Cv  Ease of use: Matlab is a relatively easy language to get to grips with. Matlab is a pretty high-level scripting language, meaning that you don’t have to worry about libraries, declaring variables, memory management or other lower-level programming issues. As such, it can be very easy to throw together some code to prototype your image processing idea. Say for example I want to read in an image from file and display it. In Matlab, you could write this  Easy, right? Now, if you wanted to do the same using OpenCV, it would look like: Virtual Teaching and Painting Platform for the Colour Blind www.iosrjournals.org 3 | Page  Memory Management: OpenCV is based on C. As such, every time you allocate a chunk of memory you will have to release it again. If you have a loop in your code where you allocate a chunk of memory in that loop and forget release it afterwards, you will get what is called a “leak”. This is where the program will use a growing amount of memory until it crashes from no remaining memory. Due to the high-level nature of Matlab, it is “smart” enough to automatically allocate and release memory in the background.  Development Environment: Matlab comes with its own development environment. For OpenCV, there is no particular IDE that you have to use. Instead, you have a choice of any C programming IDE depending on whether you are using Windows, Linux, or OS X. For Windows, Microsoft Visual Studio or NetBeans is the typical IDE used for OpenCV. In Linux, its Eclipse or NetBeans, and in OSX, we use Apple’s Xcode. IV. Algorithms Change Color 1. Click on the Change Color button. 2. Flip the video. 3. Convert the video into CMY format and finally store the grayscale forms of both formats in Gray and Cgray. 4. Find the individual color forms R,G,B,C,M,Y of the image by subtracting the grayscale forms of the image(s) from the individual color components of the same. 5. Assume a rectangular bounding box of a particular area. 6. Calculate the R,G,B,C,M,Y count for each pixel within the bounding box. 7. Calculate the sum of R,G and B counts and C,M and Y counts and store them separately. 8. Calculate the probability of each color by dividing individual pixel count of each color with the sum of RGB count(for red, green and blue only) or CMY count(for cyan, magenta and yellow only). 9. Repeat the steps 2 down to 8 for a designated no. of frames. 10. Find the color with the maximum probability and pass its id to the paint function. Virtual Teaching and Painting Platform for the Colour Blind www.iosrjournals.org 4 | Page Colour Detection Algorithm Virtual Teaching and Painting Platform for the Colour Blind www.iosrjournals.org 5 | Page Algorithm For Colour Extraction Paint 1. The id of the color with maximum probability is received. 2. The current color is set to the color with the value id. 3. Flip the video. 4. Display the video in the designated axes. 5. Find the area, centroid and bounding box properties of the black and white form of the extracted color component of the image. 6. Find the region of the image with maximum area and assume a rectangular bounding box around that region. 7. Find the centroid of the bounding box. 8. Set the value of the color components(R,G and B) of the adjoining pixels of the centroid as 255 or 0 according to the color with the value id. 9. Display the current image in the designated axes. 10. Repeat steps 7 and 8 until the status of pause button is true. 11. Repeat steps 5 down to 10 till the video is ON. 12. Flush the video and stop. Virtual Teaching and Painting Platfo

/pdf/virtual-teaching-and-painting-platform-for-the-colour-blind-57ucsnx2gy.pdf

Virtual Teaching and Painting Platform for the Colour Blind

The growing importance of huge amount of data and the data analysis for business agility has prophesied managerial decisions to ensure data security and disaster recovery options. This growing dependency has triggered extensive storage of data and development of ‘high availability’ databases. One of the prevalent practices in the industry is to provide for a data center (DC) where the on-going service runs under normal circumstances and another server at Disaster Recovery Center (DR) that contains a replica of the database to provide continuity of service if the primary database fails due to a catastrophe or some other unavoidable reasons. Such situation may cause the database server to be damaged or result in a hardware malfunction. In this paper, we propose a decentralized disaster recovery structure instead of a single centralized DR, i.e. the databases are hosted on multiple servers, where each of these servers is catering to one or more related schema. Since we have multiple servers each catering for a different schema, we also propose different replication schemes for each of these schemas depending on the nature of updates and usage. In the proposed architecture, we would like to have a cyclic back up scheme to provide better decentralization and less vulnerable as opposed to a radial back up architectures.

Distributed data recovery architecture based on schema segregation

Experiments to measure the oxidation rates of carbon by molecular and atomic oxygen were conducted using a non-thermal, radio frequency generated plasma exposed to a resistively heated graphite rod at surface temperatures between 773 and 2073 K. The temperature dependence of oxidation behavior was obtained for pure O2 and for a mixture of O2 and O atoms produced by the plasma. Oxygen atom flux to the graphite samples was characterized based on etching rates of reference polyimide films. Mass loss and exposure time of the graphite samples were measured at different test conditions to calculate oxidation rates. Mass spectrometry was used to identify the chemical species present in the oxidation products. Experiments were carried out at a pressure of 0.33 kPa which provided carbon oxidation results relevant to atmospheric entry conditions and which will be used to inform Arrhenius rate equations for gas-surface interaction computational models. 

Souvik Bhattacharya

Papers

Designing combo recharge plans for telecom subscribers using itemset mining technique

Virtual Teaching and Painting Platform for the Colour Blind

Distributed data recovery architecture based on schema segregation

High temperature oxidation of graphite under non-thermal oxygen plasma