Bankim Patel

Big Data Analysis: Recommendation System with Hadoop Framework

Recommendation system provides the facility to understand a persons taste and find new, desirable content for them automatically based on the pattern between their likes and rating of different items. In this paper, we have proposed a recommendation system for the large amount of data available on the web in the form of ratings, reviews, opinions, complaints, remarks, feedback, and comments about any item (product, event, individual and services) using Hadoop Framework. We have implemented Mahout Interfaces for analyzing the data provided by review and rating site for movies.

Web Mining: Opinion and Feedback Analysis for Educational Institutions

Big amount of data available in the forms of reviews, opinions, feedbacks, remarks, comments, observations, clarifications, and explanations that require a robust mechanism to store, retrieve, analyze, and management. In this paper, we are proposing system that provides review or summary of above mention text data available on web for an educational institute. Due to big data size above system can be extended as real time recommendation system for Big Data analysis as future enhancement.

Thermal imaging with fuzzy classifier for maturity and size based non-destructive mango (Mangifera Indica L.) grading

This is the era of ICT technologies. As its an important task to reach consumers demand for good quality mango, automation in grading of mango (Mangifera Indica L.) is required. This paper is addressing the grading issue of agricultural produce based on its maturity and size. The Fuzzy inference system is used for decision-making process. Prediction of mangos maturity is done through its skins color but for some exceptional tribe of mango like “Langdo”, skin color will remain the same for its lifetime. In such cases, normal imaging (reflective imaging) will not work for predicting its maturity. One can use infrared, x-ray or thermal imaging for maturity prediction. Here mango grading is performed based on maturity and size feature. For that, with mean intensity algorithm in L∗a∗b∗ color space and FLIR ONE thermal camera is used to predict the maturity of mango. Size of mangois predicted by three parameters namely weight, eccentricity and area. Fuzzy classifier is used for predicting size feature. To grade mango decision making theory is used and mango is graded in two different classes. Time needed to grade a mango is 2.3 seconds and accuracy received is 89%.

Dhiya: A stemmer for morphological level analysis of Gujarati language

To understand a language, analysis has to be done at word level, sentence level, context level and discourse level. Morphological analysis comes at the base of all, as it is the first step to understand a given sentence. One of the tasks that can be done at morphological level is stemming. To identify the stem term of a given word is stemming. Stemming is one of the important activities which is not just related to Natural Language Processing domain, but is equally important in Information Retrieval domain. In this paper, authors suggest DHIYA a stemmer for Gujarati language. This stemmer is based on the morphology of Gujarati language. To develop the stemmer, inflections which appeared most in Gujarati text were identified. Based on it, the rule set was created. For training and evaluation of the stemmers performance the EMILLE corpus is used. The accuracy of the stemmer is 92.41%.

A Study of Path Completion Techniques in Web Usage Mining

Path completion is a critical and difficult task in the preprocessing phase of web usage mining. We mold the data preprocessing phase to accomplish our goal to mine websites designed using a content management system (cms). The data preprocessing phase includes data cleaning, user identification, session identification, site structure and link details formation, path completion and event generation. The paper includes work on path completion by considering different types of path generated in accessing the website designed using cms and gives a novel algorithm to form the path.

Conceptual Framework for Soft Computing based Intrusion Detection to Reduce False Positive Rate

As the popularity and usage of Internet increases security concerns are also become important. Everyone want to be connected to the world through Internet protecting own resources. Intrusion Detection System is one of lucrative area for researchers since long. Numbers of researchers have worked for increasing efficiency of Intrusion Detection Systems. But still many challenges are present in modern Intrusion Detection Systems. One of the major challenges is controlling false positive rate. In this paper we have proposed Soft Computing based Intrusion Detection. We have suggested Genetic Algorithm based solution for Intrusion Detection. In place of standalone Genetic Algorithm we have proposed ensemble soft computing techniques for better results.

Shape, size and maturity features extraction with fuzzy classifier for non-destructive mango (Mangifera Indica L., cv. Kesar) grading

In the era of ICT technologies, automation in grading of mango (Mangifera Indica L.) is important to reach consumer demand for quality mango. This paper addresses that issue to identify agricultural produce based on shape, size and maturity. Fuzzy inference system is used for decision making process. In this paper, proposed methodology is divided in three phases: In first phase, mangoes are classified either well formed or deformed using eccentricity, extent and cross-ratio properties of shape. Second phase discusses about size and maturity classification. Weight and area are used for size feature extraction and mean of a* and b* channel of L*a*b* color space are used as parameters of maturity feature. In this phase, mangoes are classified in small, medium or big size and unripe, partially ripe or ripe maturity. At final phase, decision making theory is used to grade mango in class I, class II or class III. Integration of whole system results average accuracy 90% and system takes 2.1 seconds to grade a mango.

Performance Measurement and Evaluation of Pluggable to Scheduler Dynamic Load Balancing Algorithm (P2S_DLB) in Distributed Computing Environment

The imbalanced load between clusters is a key issue in distributed computing environment. All existing dynamic load balancing algorithms are post-active, as balancing activities start after system turn into imbalanced state. The better approach is to design pro-active load balancing algorithm which starts working with scheduling algorithms. It helps scheduling algorithms to schedule incoming jobs in such a way that system remains in balanced state. The pluggable to scheduler dynamic load balancing algorithm (P2S_DLB) is designed and evaluated over priority scheduling algorithm in our previous research work. The P2S_DLB is pro-active dynamic load balancing algorithm. In this paper, we have measured and evaluated the performance of P2S_DLB over First Come First Serve (FCFS), Shortest Job First (SJF), and Earliest Deadline First (EDF) scheduling algorithms. The experimental result shows that algorithm has improved the cluster utilization and decreased the imbalance level of distributed computing environment in case of all the three scheduling algorithms.

A System for the Conversion of Digital Gujarati Text-to-Speech for Visually Impaired People

In the epoch of hi-tech development, study on Text-to-Speech conversion shows remarkable enhancement in last couple of decades. Visually impaired people are not able to read, so Text-to-Speech system acts as an aid for visually impaired people for reading by hearing the text. In this paper, we presented the development of computer-based Gujarati Text-to-Speech system that delivers text in Gujarati audio form. Arbitrary digital Gujarati text is considered as an input to the system; conversion is done with regard to the Akshara of Gujarati language, and sound is produced in the form of phoneme, diphone, or syllable as per the requirement. Single audio file is created of the text so that it can also be heard at later stage. The detailed algorithm along with the format of speech database is also presented in the paper. Proposed system is tested on the documents collected from online news Web site and it gives satisfactory result.

Transliteration of Digital Gujarati Text Into Printable Braille

In the era of technological development, the enhancement is witnessed in the way Braille is produced, stored and retrieved, which makes it easier to use at different places. But it is still difficult to produce an error-free Braille for multifaceted scripts. In this paper, the deliberation is done on the development of translation tool that converts Gujarati text along with English and Hindi text into Braille and store it into a data file that can be directly printed on the Embosser. The technique is explicated by keeping in wits the perspective of the standard Braille system and the table for transformation is also specified. The system gives satisfactory results. As very limited amount of literature is available for visually impaired people of Gujarat, so the proposed system will increase literature in Braille which can be used by them. This paper also forwards the new thought for the assistance to the visually impaired people and Braille learners.