Search

Statistical machine translation (SMT) is a variant of machine translation where the translations are handled with statistically defined rules. Numerous researchers have attempted to build the framework which can comprehend the different dialects to translate from one source language to another target language. However, the focus on the translation of poetry is less. Reliable and rapid transliteration of the poetry is very mandatory for the execution of the computer to translate the poem from one language to another. The existing approach has several issues such as time consumption, quality of the translation process, and matching of similar words. To overcome these issues, we propose a phrase-based statistical machine translation (PSMT) with special adherence to word sense disambiguation (WSD). Quality of the translation is increased by sensing the ambiguous words with WSD. The Hindi WordNet along with the Lesk algorithm identifies the ambiguous words and senses the exact meaning before the phrase extraction. Finally, the proposed method is compared with machine translation schemes such as rule-based machine translation and transfer-based machine translation. The experimental results suggest that the proposed method performed well with the inclusion of WSD in the PSMT technique.

Objectives: To provide approaches for effective Hindi-to-English Machine Translation (MT) that can be helpful in inexpensive and ease implementation of and MT systems. Methods/Statistical Analysis: Structure of the Hindi and English languages have been studied thoroughly. The possible steps towards the Natural languages have also been studied. The methods, rules, approaches, tools, resources etc. related to MT have been discussed in detail. Findings: MT is an idea for automatic translation of a language. India is the country with full of diversity in culture and languages. More than 20 regional languages are spoken along with several dialects. Hindi is a widely spoken language in all the states of country. A lot of literature, poetries and valuable texts are available in Hindi which gives opportunities to retranslate into English. However, new generation is learning English rapidly and also showing keenness to learn it in simplified lucid manner. Several efforts have been made in this direction. A large number of approaches and solutions exist for MT still there is a huge scope. The paper addresses the challenges of MT and solution efforts made in this direction. This motivates researchers to implement new Hindi-to-English Machine translation systems. Application/Improvements: Efficient, inexpensive and ease translation for available Hindi literature, poetries and other valuable texts into English. Children can easily learn the culture through the poetries and literatures hence the Machine Translation of these will bring wonderful impact.

We are a part of that society which has different-different cultures, languages, rituals, etc. There are some sentences which we used to say someone at particular situation or to motivate someone which we called as couplets. A couplet is a pair of consecutive lines of rhythm in poetry. Couplet usually consists of two consecutive lines that poem has the same rhythm. Couplets are balanced in length and word segmentation, in a manner that matching words in the two consecutive sentences match each other by observing certain limitation on semantic, syntactic, and lexical connection. In this research, authors have tried to translate few couplets from English to Hindi language and this kind of task comes under the umbrella of natural language processing. In continuation of the same, authors have used a statistical machine translation (SMT) approach for producing good results (of English to Hindi translation of couplets).

Web usage mining is the application of data mining techniques and is used to extract the important data which are present in the web. Nowadays web log mining is a very popular and computationally expensive task. Preprocessing, pattern discovery, and pattern analysis are the major task of web usage mining. In this paper we are presenting an overview of existing algorithms used in pattern discovery phase for mining the frequent item set by designing comparative analysis table i.e. Apriori, K-Apriori, FP growth which are used in pattern discovery phase. General Terms Web usage mining, Frequent item set mining.

present word web is huge storage of information and it will keep increasing with growing of internet technologies. But the human being capability to read, access and understand content does not increase with that tread. Hence it becomes complex to website owners to present proper information to the users. This led to provide personalized web services to users. One of the well-known approaches in providing web personalization is Web Usage Mining. In this paper, our motive of web usage mining is to discover users' access patterns of web pages automatically and quickly from the huge sever access log records, such as frequently visited hyperlinks, frequently accessed web pages and users grouping. Also, we proposed a new method for discovering users' access patterns and recommend it to the user.

Idiom plays a significant role in a language literature (Sinha in A system for identification of idioms in Hindi. IEEE, 2014 [1]). It enhances the glory of the language which contains it. As idiom points to a figurative meaning (Anastasiou in Idiom treatment experiments in machine translation, 2010 [2]), i.e., the meaning different from the word set present in that idiom, it becomes difficult for the machines to translate them. This paper proposes an achievable implementation way of Hindi to English idiom translation using a hybrid approach which consists of interlingual-based approach combined with the transfer-based approach. This paper also provides details on system requirements, algorithm, database connection, and result analysis.

Nowadays, the artificial intelligence plays an important role in technological aspects. AI uses big data for performing its job, therefore it is necessary to store, analyze, and control the big data. AI needs to perform works like face detection and image classification. The classification process categorizes small pixels that are present inside a digital image into several different classes. Generally for classification, multispectral data were used by obtaining the spectral pattern present between image data and the numerical basis for this categorization are pixels. We propose an open-source Hadoop image classification platform. The goal of this research is to make a product that will, in future, be used to make image classification and processing tool on extensive number of images which will empower scholars and scientists to create applications.

Web usage mining is a task of data mining of web data that extract knowledge about web users and how these users interacting with the web site. Our motive of web usage mining is to provide web intelligence by discovering users' access patterns of web pages, such as frequently visited hyperlinks, frequently accessed web pages, popular access sequence of web pages, and users grouping etc., automatically and quickly from the huge sever access log records. Through web usage mining, we can mine the server log data, registration information of users and other relative information left by user access. This data will provide base for analysis which helps the organization to make decisions and to personalize the web pages.

In large databases where continuous increase in the volume and detail of data captured by organizations [1] and accessed by different users and at different access levels, need to incorporate the security and transparency during the different kinds of information extraction according to role based access levels. Therefore in this paper an efficient data model for improving the privacy as well as transparency of a data model is proposed. The proposed data model is implemented on the different access levels and a cryptographic security is implemented during data access. Additionally a case study on the governmental organization for distribution of food and fertilizers are reported. The implementation of the proposed security and access control model is provided using JAVA technology. After implementation the performance of the system is evaluated and their performance is reported. The results show the model is efficient and able to prevent the data access for unauthorized access.

Online email chronicles are an under-ensured yet greatly delicate data asset. Email documents can store year of individual and business email in a simple to-get to structure, one that is much less demanding to trade off than messages being transmitted on the wire. Most email files, be that as it may, are secured by reusable passwords that are frequently frail and can be effortlessly bargained. To secure such files, we propose novel multilevel email security building design. The proposed structural planning deals with three levels of security which are picture confirmation through example matching, pressure & cryptography in light of characteristic. At the starting levels our methodology is by all accounts at more elevated amount that the current ones. It is extraordinary that moderate mediums that course messages in the middle of sender and beneficiaries can be a genuine risk to security as these halfway can be effortlessly catch and messed with email messages numerous programming based arrangements has been proposed to tackle these issue, for example, it were created however these arrangements were sufficiently bad to give security and different assaults can meant to it and they can misuse the vulnerabilities of these administration [2]. It is vital to forestall such phishing assaults. One of the approaches to keep the watchword burglary is to abstain from utilizing passwords and to confirm a client without a content secret key. This work proposes a safe confirmation administration construction modelling ISA-CC (Image Sequence Authentication-Compression & Cryptography) that is picture based and wipes out the requirement for content passwords. [1]

In cloud computing, at storage level, security provided to cloud by access control and data encryption key, and at network level installing virtual firewall and cloud service provider provides its internal storage for storing client's data using ips/ids to protect against attacks. Storing data in encrypted format is a common method of data privacy protection which causes a risk for the unauthorized disclosure of the user data. In this paper, discuss the introduction of cloud computing, framework for public cloud security, cloud service providers, authentication mechanism and attributes. There all different mechanism is explained for providing privacy in public cloud services on basis of the desired and proposed architecture.

With Easy and fast deployment of Mobile Ad-hoc Networks (MANETs), the capability of MANET to support real time services with constricted quality of service requirements has come ahead. The IEEE 802.11e Enhanced Distributed Channel Access (EDCA) mechanism concede prioritized medium access for applications with high QoS requirements by assigning different priorities to its four access categories. In this paper we expose results relative to the interaction of AODV routing protocol for MANETs and the IEEE 802.11e EDCA mechanism. The performance is evaluated under real time audio traffic through simulation using OPNET modeler 14.0. The performance of protocol is measured in terms of various quality of service measures including hops per route, route discovery time, delay and throughput are calculated and graphs has been plotted. Simulation results show that very substantial improvements in terms of AODV performance parameters and minimum delay are achieved due to increased routing responsiveness.

There has been a significant growth in the use of mobile ad hoc network to support network applications. Real time applications are more exposed to delays and as such proper queuing and traffic management at the packet level must be considered to provide the required quality of service to the end user. Queuing mechanisms have been a subject of demanding discussion and research in the network field for scheduling packets from different traffic flows for processing at a network node (Gateway). Various queue scheduling mechanisms have been designed to improve the quality of service for networked multimedia applications. In this paper, we estimate the performance of four queuing mechanisms (FIFO, PQ, DWRR and MDRR) that are widely deployed for differentiated services using a hypothetical network topology. In this paper we carry out simulation using OPNET and compare their relative performance based on AODV routing protocol. The performance metrics used are throughput and queuing delay.