APPLICATION OF SMART IRRIGATION SYSTEM
Authors: Ashish Kumar Parashar
Abstract
The major aim of the present research is to build up technique in agriculture automation to take over, in the present scenario and deliver to its full potential. This system is designed by using arduino microcontroller, to overcome limitations of agriculture farming to supply water to plants by furrow and sprinkler methods with the available water tables. In this study an UNO Arduino microcontroller is used, which receives the input signal of the changing moisture state of the soil through the moisture sensors. On receiving the signal by the microcontroller, an output is generated for driving the relay to operate the water pump, simultaneously indicating the moisture content of the soil of the field, on mobile or laptop through Bluetooth. A buzzer is also interfaced with controller to warn, when the moisture content goes below the minimum limit and when it goes above the maximum limit. The observations of moisture content are taken for three crops, Mustard, Wheat and Rice and the field is irrigated by Furrow and sprinkler methods. The outcome of the present study indicates that Desired Moisture Content (DMC) of 23.03%, 20.22% and 30.05% for Mustard, Wheat and Rice crops is achieved with smart irrigation system using furrow and sprinkler methods. This smart irrigation system developed in the research is found to function with an overall accuracy of 98 %.
Using unstructured medical Notes for patient long-term health prognosis
Authors: M Lavanya, D Meghana, G Swarna Latha, T Amareswari Preethi, B Prathyusha, M Sreeja
Abstract
Patient data are a critical task in diverting priority attention to precise risk classification. The most advanced versions rely mostly on the digitization of electronic health records in organised form (EHRs). Such models miss the useful patient information incorporated in unstructured clinical notes, which is the predominant medium used by carers to document the illness chronology of patients. The availability of such patient-specific data provides an unparalleled chance to develop intelligent systems that offer a unique insight into the physiology of patients’ illness. In addition, only very few research attempted to measure the performance of deep neural networks in publicly available data sets versus state of the art models. This study provides important findings on the applicability of deep learning models for the clinical task of ICD-9 group prediction from our benchmarking tests. We offer FarSight, a long-term aggregation method designed to spot the early signs of the illness. The semantine information is taken from the patient representations by the use of vector space and topic modelling techniques. MIMIC-III database experiments highlighted higher performance of suggested unbuilt-data models, which improved by 19.34 percent in AUPRC and 5.41 percent in AUROC when compared to a structured EHR-based state of the art model in AUPRC
Influence of Activity Minimization and Misinformation in Social Networks
Authors: Shaik Mahaboob Basha, N Naveen, J Venkatesh, K Sai Krishna, K Tirumala Krishna
Abstract
The widespread online misinformation could cause public frenzy and genuine monetary harms. The misinformation containment issue aims at limiting the spread of misinformation in online interpersonal organizations by dispatching contending efforts. Roused by sensible situations, we present an investigation of the misinformation containment issue for the situation when a self- assertive number of falls are permitted. This paper makes four commitments. In the first place, we give a formal model for multi-course dissemination and present an important idea called as course need. Second, we show that the misinformation containment issue can’t be approximated. Third, we present a few kinds of course need that are as often as possible found in genuine informal communities. At long last, we plan novel calculations for tackling the misinformation containment issue. The viability of the proposed calculation is upheld by empowering experimental outcomes