Entity

Time filter

Source Type

Karnataka, India

Mehrotra R.,BITS Pilani | Agrawal R.,BITS Pilani | Haider S.A.,MIT Manipal
ACM International Conference Proceeding Series | Year: 2012

Machine Learning algorithms are often as good as the data they can learn from. Enormous amount of unlabeled data is readily available and the ability to efficiently use such amount of unlabeled data holds a significant promise in terms of increasing the performance of various learning tasks. We consider the task of supervised Domain Adaptation and present a Self-Taught learning based framework which makes use of the K-SVD algorithm for learning sparse representation of data in an unsupervised manner. To the best of our knowledge this is the first work that integrates K-SVD algorithm into the self-taught learning framework. The K-SVD algorithm iteratively alternates between sparse coding of the instances based on the current dictionary and a process of updating/adapting the dictionary to better fit the data so as to achieve a sparse representation under strict sparsity constraints. Using the learnt dictionary, a rich feature representation of the few labeled instances is obtained which is fed to a classifier along with class labels to build the model. We evaluate our framework on the task of domain adaptation for sentiment classification. Both self-domain (requiring very few domain-specific training instances) and cross-domain classification (requiring 0 labeled instances of target domain and very few labeled instances of source domain) are performed. Empirical comparisons of self-domain and cross-domain results establish the efficacy of the proposed framework. © 2012 ACM.


Ibrahim S.,Louisiana Tech University | Chowriappa P.,Louisiana Tech University | Dua S.,Louisiana Tech University | Acharya U.R.,Ngee Ann Polytechnic | And 3 more authors.
Medical and Biological Engineering and Computing | Year: 2015

Prolonged diabetes retinopathy leads to diabetes maculopathy, which causes gradual and irreversible loss of vision. It is important for physicians to have a decision system that detects the early symptoms of the disease. This can be achieved by building a classification model using machine learning algorithms. Fuzzy logic classifiers group data elements with a degree of membership in multiple classes by defining membership functions for each attribute. Various methods have been proposed to determine the partitioning of membership functions in a fuzzy logic inference system. A clustering method partitions the membership functions by grouping data that have high similarity into clusters, while an equalized universe method partitions data into predefined equal clusters. The distribution of each attribute determines its partitioning as fine or coarse. A simple grid partitioning partitions each attribute equally and is therefore not effective in handling varying distribution amongst the attributes. A data-adaptive method uses a data frequency-driven approach to partition each attribute based on the distribution of data in that attribute. A data-adaptive neuro-fuzzy inference system creates corresponding rules for both finely distributed and coarsely distributed attributes. This method produced more useful rules and a more effective classification system. We obtained an overall accuracy of 98.55 %. © 2015, International Federation for Medical and Biological Engineering.


Santhosh K.V.,MIT Manipal | Roy B.K.,National Institute of Technology Silchar
International Journal of Bio-Science and Bio-Technology | Year: 2015

This paper aims at designing an oxygen level monitoring technique in an oxygen cylinder. The amount of oxygen present inside an oxygen cylinder is a very vital information when such cylinder is in use as life saving measure to a critical patient. In this paper, it is proposed to measure oxygen level using pressure and temperature sensors. Conditioned output of these sensors is connected as input to ARM micro-controller. ARM is programmed to display the actual pressure of oxygen cylinder in terms of numerical values and also in terms of fuzzy variables. A buzzer is also used as indicator to caution the attendants of patients whenever the level of oxygen is below a pre-decided value. The signal from the cylinder is further transmitted to the monitoring station through wireless communication module. Graphical display is used at monitoring station to indicate pressure of all oxygen cylinders to initiate actions like use cylinders which are in good conditions, replacement of empty cylinders with filled ones, etc. Experimental results show that, the aims set for this work are achieved satisfactorily. © 2015 SERSC.


Ravindra B.V.,Manipal University India | Sriraam N.,M.S. Ramaiah Institute of Technology | Geetha M.,MIT Manipal
Proceedings of International Conference on Circuits, Communication, Control and Computing, I4C 2014 | Year: 2014

The contributing factors for kidney dialysis such as creatinine, sodium, urea plays an important role in deciding the survival prediction of the patients as well as the need for undergoing kidney transplantation. Several attempts have been made to derive automated decision making procedure for earlier prediction. This preliminary study investigates the importance of clustering technique for identifying the influence of kidney dialysis parameters. A simple K-means algorithm is used to elicit knowledge about the interaction between many of these measured parameters and patient survival. The clustering procedure predicts the survival period of the patients who is undergoing the dialysis procedure. © 2014 IEEE.


Samaga R.L.,National Institute of Technology Karnataka | Vittal K.P.,National Institute of Technology Karnataka | Vikas J,MIT Manipal
2011 IEEE PES International Conference on Innovative Smart Grid Technologies-India, ISGT India 2011 | Year: 2011

Condition monitoring units are employed in industries to monitor the health of the machines continuously. Air gap eccentricity fault is one of the asymmetrical faults which can result in the machine failure. Motor Current Signature Analysis and Vibration Analysis are the two most popular methods used for eccentricity fault detection in the induction motor. In this paper, a study conducted on an induction motor to analyse the effect of supply voltage unbalance on the method of eccentricity fault detection by Motor Current Signature Analysis is presented. A dynamic model of the induction motor suffering from air gap eccentricity and has the capability to take unbalance supply voltage is developed and the results obtained by simulating this model are validated by the experiments conducted on an induction motor suffering from inclined mixed eccentricity and fed with unbalance voltage supply. © 2011 IEEE.

Discover hidden collaborations