G. Santhosh Kumar

Using open source software for digital libraries A case study of CUSAT

Purpose – The purpose of this paper is to describe the design and development of a digital library at Cochin University of Science and Technology (CUSAT), India, using DSpace open source software. The study covers the structure, contents and usage of CUSAT digital library.Design/methodology/approach – This paper examines the possibilities of applying open source in libraries. An evaluative approach is carried out to explore the features of the CUSAT digital library. The Google Analytics service is employed to measure the amount of use of digital library by users across the world.Findings – CUSAT has successfully applied DSpace open source software for building a digital library. The digital library has had visits from 78 countries, with the major share from India. The distribution of documents in the digital library is uneven. Past exam question papers share the major part of the collection. The number of research papers, articles and rare documents is less.Originality/value – The study is the first of it...

Virtualization Techniques: A Methodical Review of XEN and KVM

Over the past few years, researchers have been propelled for a Utility Computing Model. Cloud computing allows delivering resource on demand by means of virtualization. Virtualization has been around from the period of Mainframe computing. The proposal of using a computer system to emulate another computer system was early realized as useful for testing and increased resource utilization purposes. As with several computer technologies, IBM initiated the way with their VM system. In the last decennary, VMware’s virtual machine monitor has been quite successful. In recent times, open-source hypervisor’s like Xen and KVM added virtualization to the open source world, initially with a variant named para-virtualization and later using hardware assisted full virtualization. This paper surveys two main virtualization technologies: Xen and KVM. Also system and network performance evaluation tests are conducted to analyze scalability and performance of the virtualized environment.

English to Malayalam translation: a statistical approach

This paper underlines a methodology for translating text from English into the Dravidian language, Malayalam using statistical models. By using a monolingual Malayalam corpus and a bilingual English/Malayalam corpus in the training phase, the machine automatically generates Malayalam translations of English sentences. This paper also discusses a technique to improve the alignment model by incorporating the parts of speech information into the bilingual corpus. Removing the insignificant alignments from the sentence pairs by this approach has ensured better training results. Pre-processing techniques like suffix separation from the Malayalam corpus and stop word elimination from the bilingual corpus also proved to be effective in training. Various handcrafted rules designed for the suffix separation process which can be used as a guideline in implementing suffix separation in Malayalam language are also presented in this paper. The structural difference between the English Malayalam pair is resolved in the decoder by applying the order conversion rules. Experiments conducted on a sample corpus have generated reasonably good Malayalam translations and the results are verified with F measure, BLEU and WER evaluation metrics.

Association rule extraction from medical transcripts of diabetic patients

Medical databases serve as rich knowledge sources for effective medical diagnosis. Recent advances in medical technology and extensive usage of electronic medical record systems, helps in massive production of medical text data in hospitals and other health institutions. Most of this text data that contain valuable information are just filed and not utilized to the full extent. Proper usage of medical information can bring about tremendous changes in medical field. This paper present a new method of uncovering valid association rules from medical transcripts. The extracted rules describes association of disease with other diseases, symptoms of a particular disease, medications used for treating diseases, the most prominent age group of patients for developing a particular disease. NLP (Natural Language Processing) tools were combined with data mining algorithms (Apriori algorithm and FP-Growth algorithm) for the extraction of rules. Interesting rules were selected using the correlation measure, lift.

Face recognition using probabilistic neural networks

In this paper we address the problem of face detection and recognition of grey scale frontal view images. We propose a face recognition system based on probabilistic neural networks (PNN) architecture. The system is implemented using voronoi/ delaunay tessellations and template matching. Images are segmented successfully into homogeneous regions by virtue of voronoi diagram properties. Face verification is achieved using matching scores computed by correlating edge gradients of reference images. The advantage of classification using PNN models is its short training time. The correlation based template matching guarantees good classification results.

Automatic image annotation using SURF descriptors

In recent years there is an apparent shift in research from content based image retrieval (CBIR) to automatic image annotation in order to bridge the gap between low level features and high level semantics of images. Automatic Image Annotation (AIA) techniques facilitate extraction of high level semantic concepts from images by machine learning techniques. Many AIA techniques use feature analysis as the first step to identify the objects in the image. However, the high dimensional image features make the performance of the system worse. This paper describes and evaluates an automatic image annotation framework which uses SURF descriptors to select right number of features and right features for annotation. The proposed framework uses a hybrid approach in which k-means clustering is used in the training phase and fuzzy K-NN classification in the annotation phase. The performance of the system is evaluated using standard metrics.

RMAC-M: Extending the R-MAC protocol for an energy efficient, delay tolerant underwater acoustic sensor network application with a mobile data mule node

One of the major applications of underwater acoustic sensor networks (UWASN) is ocean environment monitoring. Employing data mules is an energy efficient way of data collection from the underwater sensor nodes in such a network. A data mule node such as an autonomous underwater vehicle (AUV) periodically visits the stationary nodes to download data. By conserving the power required for data transmission over long distances to a remote data sink, this approach extends the network life time. In this paper we propose a new MAC protocol to support a single mobile data mule node to collect the data sensed by the sensor nodes in periodic runs through the network. In this approach, the nodes need to perform only short distance, single hop transmission to the data mule. The protocol design discussed in this paper is motivated to support such an application. The proposed protocol is a hybrid protocol, which employs a combination of schedule based access among the stationary nodes along with handshake based access to support mobile data mules. The new protocol, RMAC-M is developed as an extension to the energy efficient MAC protocol R-MAC by extending the slot time of R-MAC to include a contention part for a hand shake based data transfer. The mobile node makes use of a beacon to signal its presence to all the nearby nodes, which can then handshake with the mobile node for data transfer. Simulation results show that the new protocol provides efficient support for a mobile data mule node while preserving the advantages of R-MAC such as energy efficiency and fairness.

Alignment Model and Training Technique in SMT from English to Malayalam

This paper investigates certain methods of training adopted in the Statistical Machine Translator (SMT) from English to Malayalam. In English Malayalam SMT, the word to word translation is determined by training the parallel corpus. Our primary goal is to improve the alignment model by reducing the number of possible alignments of all sentence pairs present in the bilingual corpus. Incorporating morphological information into the parallel corpus with the help of the parts of speech tagger has brought around better training results with improved accuracy.

An adaptive cluster based routing scheme for mobile wireless sensor networks

Clustering schemes improve energy efficiency of wireless sensor networks. The inclusion of mobility as a new criterion for the cluster creation and maintenance adds new challenges for these clustering schemes. Cluster formation and cluster head selection is done on a stochastic basis for most of the algorithms. In this paper we introduce a cluster formation and routing algorithm based on a mobility factor. The proposed algorithm is compared with LEACH-M protocol based on metrics viz. number of cluster head transitions, average residual energy, number of alive nodes and number of messages lost.

An effective private data storage and retrieval system using secret sharing scheme based on secure multi-party computation

Privacy of the outsourced data is one of the major challenge. Insecurity of the network environment and untrust-worthiness of the service providers are obstacles of making the database as a service. Privacy concerns exist wherever personally identifiable information is collected and stored. Public databases and resources are potential source of risk to user privacy. An intentional database owner can guess user details by practically monitoring their queries. Hence the major challenge here is to share the data by protecting personal information. A data retrieval scheme allowing the users to query the database with out compromising the privacy in data item is generally sought The naive solution for confidentiality is to encrypt data before outsourcing. Query execution, key management and statistical inference are major challenges in this case. The proposed system suggests a mechanism to store private information and secure retrieval of this data using secret sharing based Secure Multiparty Computation(SMC). The idea is to develop a mechanism to store private information with a highly available storage provider which could be accessed from anywhere using queries while hiding the actual data values from the storage provider. Multiparty Computation facilitates application of join functions over their private inputs and SMC performs these functions by keeping the input data private. This is achieved by making secret shares of the inputs and manipulating the shares to compute some functions.