Search

Identifying and curtailing the spread of fake news is a complex and challenging task. Automated stance detection can be an important first step in alleviating the tedious process of fact checking. In this paper, we try to evaluate various language representation, and statistical similarity technique to discern the best possible mathematical modeling of the sentences for the task of stance evaluation. Our work is based on the dataset of the fake news challenge 1. The paper expounds our implementation details, the features, and models that work best for the given task.

The protection of crops is very important for the improved quality of agricultural production. It plays an important role in mankind and the environment. It is significant to protect the crops from pests and plant diseases that harm the crops. Proper and accurate information about the plant/crop and soil while growing can make the crop healthy and also prevent from serious harm that may occur in the future, by using pesticides and other prevention techniques. Yield and crop production can be escalated by implementing optimum methods. With the advancement of science and technology, the growing work in this domain using sensors and other IoT equipment will be a boon for the society and environment. This research paper focuses to explore different IoT-based techniques that are available or are being researched upon for protection of crops. The paper presents an IOT-based anti-noise technique that on the protection of plants by using several techniques that bring out the best possible results in crop protection.

Smart phones have come all the way up to become an essential part of our lives compared to any other device. People use it during their free time, some people work on their smart phones, for social media, photos, retrieving information, etc. As it can be utilized in many different fields, human body medical and fitness are also a very important field that smart phones can be used to simplify the process of medication and human body surveillance. Imagine a mobile application constantly analyzing the movement of the body through the help of sensors present in our smart phone in real time, and looking through these values, it can detect the current state of a human body like walking, sitting, laying, standing, etc., altogether human activities. In this paper, we aim to test and analyze various machine learning algorithms such a way that it recognizes the current state of the body and predicts the activity being performed with maximum accuracy. The trained models can further be deployed for real-time use in either mobile devices or on computer using a website as interface. Most of the smart phone contains accelerometer and gyroscope which are the sensors built-in with the smart phone hardware. We have taken the dataset from the UCI machine learning repository.

With the data growing in abundance amount and usage of social media increasing day by day, the features of Youtube videos, video streaming apps, e-commerce sites etc. have been an important concern for the companies. Every user wants the recommendation according to his or her choice. Therefore, it becomes very important for these applications and sites to have an efficient recommender system that not only recommends the result as per the user but also do not overlay excessive burden on the applications. In this paper, we would be analyzing the collaborative-based recommender system algorithms like (Pearson, cosine and matrix factorization) that are used under the different models of collaborative-based recommender systems. Following on the basis of which, we have proposed a generic book recommender system that works under the collaborative-based architecture.

Trend analysis is the method of analyzing or predicting the future of the goods that are available with seller. It gives the various factors affecting the performance of the store based on the past data. The research provides a detailed insight of statistics of one of the leading retail chains in London having more than 15 stores which sell laptops and accessories and investigates the current status of the company. The research aims at a thorough understanding of the customer relationship with the company. Data mining has emerged as an interdisciplinary domain in the process of knowledge discovery in databases in which techniques are applied in order to extract patterns. Further, advanced approach of analytics in order to discover and communicate meaningful patterns in data has, as a result, led to drive smart decisions. This research discusses the behavior of customer with respect to different stores in three major steps; they include (i) collection of dataset, (ii) preprocessing of the dataset, and (iii) descriptive analysis. Each step plays a key role in deriving the results. The collection of data further affects the data quality. Thus, first step ensures that the data collected is relevant and least erroneous. The approach in the report has enabled and high lightened a detailed insight in understanding trends among customers (understanding trends), i.e., descriptive analysis like dependency of sales on various factors.

The focus of this chapter is to design a cognitive information retrieval (CIR) framework using inference engine (IE). IE permits one to analyze the central concepts of information retrieval: information, information needs, and relevance. The aim is to propose an inference engine in which adequate user preferences are considered. As the cognitive inference engine (CIE) approach is involved, the complex inquiries are required to return more important outcomes as opposed to customary database questions which get irrelevant and unsolicited responses or results. The chapter highlights the framework of a cognitive rule-based engine in which preference queries are dealt with while keeping in mind the intention of the user, their performance, and optimization.

In the modern era of e-commerce and social networking, people wants to get everything best and better from every other person in their group. There is always an urge within people to find products, services and opportunities better than everyone else. Nowadays, business grows by advertising it online because people are always online and they find it more interesting to review products on a social platform. Mining of association rules is an advanced data mining research area. Currently many works are undergoing in this domain. Completeness is one of the key strengths of fuzzy association rule mining. Fuzzy cannot work with large datasets. The massive amount of candidate item sets sometimes makes it unproductive for a data mining system to analyze them. At the end, it produces a significant number of fuzzy associations. The research focuses on helping the needful to find the best suited movies depending on the previous ratings. Rules have been generated by using fuzzy association rule mining.

Green computing is one of the emerging aspect of technology that focuses on developing computer software and hardware devices which helps to reduce the power usage and carbon consumption. Today, with proliferation of smartphones, tablets and other unwired gadgets, efficient use of energy is becoming a key consideration while designing any software or hardware component. Manufacturers and developers look forward to energy-efficient solutions across the stack, with more established results through research and innovations in application design, hardware/architecture, operating systems, and runtime systems. The aim of the research is to evaluate the recall and precision of the dataset using the various Divergence from Randomness (DFR) models. The corpus used in the research was a Green Computing dataset that consisted of about 300 questions along with their answers. The topic files and Query Relevance judgement (QREL) files were created for all the questions and the answers. The results indicate that OKAPI probabilistic model has the highest precision.

Usability is a measure of determining the ease of use of an application. It is for sure an exceptionally crucial part from a client’s and business point of view. This paper means to discover strategies to improve the ease of use of an application. It utilizes the CW strategies to delineate user’s viewpoint to the applications usability criteria. It familiarizes with how an application is seen by individuals of fluctuating age. The primary thought is to check the ease of use of an application and upgrade it, if required, such that it serves useful for all scope of clients.

Relevance feedback is an essential feature that improves the final retrieved results. The working efficiency of the relevance feedback depends on feedback information given by the user each time the system gets some retrieved set of results. In order to improve the relevance feedback information, we have applied algorithms on the dataset and prepared a model. Sentiment analysis is further applied on the dataset to transform it into an accurate form and then predict the nature that has been classified into three: positive, negative or neutral sentiments. Decision tree model is used for the dataset after which it is further processed for testing and validation. In the last, a model is prepared that can, not only justify the sentiments on the basis of positive, negative and neutral but also calculates the accuracy of analyzing the statements which can act as a great help for relevance feedback systems.

Information retrieval is a process of representing, retrieving and normalising data items. The retrieval system is a method that verifies how a system responds against users’ needs. The accessing of useful information is directly related by the user’s job and the conceptual view of the information possessed by the retrieval system. In order to increase the efficiency of the retrieval system, the authors have considered new fields (TITLE and DESC) for evaluating the recall and precision parameters. This paper demonstrates the comparison of baseline probabilistic models with document fields in the retrieval process and for experimental analysis. The authors’ have used the standard TREC Ad hoc test collections based on the weighting and field models.

This is a study done for the social cause that was increasing at an alarming rate and was creating a situation of panic among the people of the world, Mortality Rate. This situation was analyzed by analyzing various factors such as birth rate, literacy rate, number of health centers, etc. using the decision tree technique in R tool which illustrated trees of two different decades separately and analyzed the factors affecting the mortality rate with their contribution in driving its rate, and also the summary of decision tree will indicate its accuracy and kappa factor to judge the authenticity of the factors chosen. This will be useful to the governing bodies to get to know about the factors and work upon them for the decrement of the infant mortality rate.

Sentiment analysis basically consist of determining and categorizing opinions expressed on the basis of computational text. It helps us to determine user’s attitude towards a specific topic on the basis of positive, neutral and negative. Social media acts as one of the biggest platforms where people express their feelings on variety of issues by writing on websites, blogs, public forums or online groups. Sports nowadays have also become one of the major topics that people discuss on social media. People’s opinions are basically positive, neutral or negative, so only looking at each opinion and coming to a result is very difficult and time consuming. In this research paper, an overall process of feature based sentiment analysis is showcased and opinions of people are analyzed regarding cricket matches using RapidMiner.

The paper reviews the research of Information Retrieval models of DFR. Tipster Ad hoc Association Press newswire retrieval task is mainly applied in IR systems. The evaluation of the AP (Associated Press) newswire is based on the values obtained in Mean Average Precision (MAP). The purpose of the paper is to come up with the most optimized DFR model for the newswire so that the user can retrieve best results according to their queries. The results reveal that a higher performance of the probabilistic DFR models is achieved in contrast to other IR models.

The paper explores the hybrid of SVM and system of Artificial Neural Network as the finest binary classification system for calculating the diabetic nature of people in comparison to Support Vector Machine (SVM). In this research, the sets of all the parameters describing the diabetic nature of a person are taken from the laboratories. This approach is chosen so that a better learning method can be used for various problems. The testing results were found to be in accordance with the accepted results that resemble with the direct diagnosis of a physician. The results of this research shows that this hybrid SVM and Artificial Neural Network (ANN) model is more precise than the SVM model. These results of the hybrid SVM and ANN model suggest that it is very effective for the classification of Diabetic and Non Diabetic nature of a person. This paper highlights the concept of Support Vector Machines and its integration with Artificial Neural Network, the two key characteristics with one being the generalization theory of the SVM model that best describes how to select a hypothesis and functions given by Kernel that introduces the idea of non-linearity without the inclusion of the actual algorithm.

The paper investigates various factors (System Reliability, User Efficacy, User behavior and User cognitive skills) that affect a user while retrieving information. These factors can be used to improve the efficiency of the system. In this research, we have created a model improving the classical IR model. The factors that affect the user are clubbed together to form a User Characteristic Data (UCD). The system interacts with the UCD to obtain the results as per the user query. The model takes in user's relevance feedback to update the UCD. The objective here is to retrieve better results every time the user wants some information. Therefore, the IIR models should consider these factors while evaluating the documents for the query given by the user for his information need.

In recent years, with the rapid increase in Internet usage, the data that has been generated is huge and unstructured. These data can be interpreted with various techniques of Data Mining. Many useful patterns can be extracted from these trends. Classifying these data into meaningful analysis is the key concept behind this study. In this paper, the authorship data for books was used. A data was created where various attributes of users were stored along with the book that they like to read. Naive Bayes was applied on the data set to find which factor is majorly affecting the ratings of the books. The various attributes were compared using data mining tool and found that the rating of books highly depends upon the location of the user. This interpretation was also verified by the measure of precision and recall. High precision results into more accuracy of the system.

Carbon plays an essential role in the environment for climate change. The presence and absence of carbon directly affects all living beings. Trees inhale carbon for giving us oxygen. The environmental study of carbon is a major concern these days. Carbon Dioxide is stored in different five carbon pools of forest. Many countries are innolved in the research of environmental factors these days. The focus of this paper is to build a system using Naive Bayes Approach that trains a model to classify forest on the basis of carbon stock and predict the level of carbon stock in the forest. The model is validated using dataset of the previous year data.

In recent years, the rapid development of Internet of Technology (IOT) makes the intelligent home come true as people expect. The intelligent home system creates the more comfortable, safer, humane and intelligent living environment. It can resolve the problems facing by the people who have busy schedules and get a very less amount of time to spend at home which is increasing rapidly around the world. For the solution of this problem, user can depend on the automated machines and gadgets like smart phones. These smart gadgets are using cloud computing which sends and receives signal o the cloud. The data that is of our use can be fetched by matching some key values using the concept of information retrieval. The key objective of this paper is to create a full-fledged application which could let user to operate the lights of their house from any remote location. The user have a list of options to select which light is to be on and when. The only requirement is to have working wifi at home to which the lights are connected. The is developed in Lua Language by using the Esplorer Integrated Development Environment (IDE). We have also used the micro controller chip ESP 8266 to build our board.

Mining right patents from database have always been interesting and the most difficult task for analysis purposes. The choice of right data mining tool and algorithm is requisite for reducing the search space, thus enabling extraction of meaningful and useful information for technology forecasting. With clustering approach, this can be easily achieved. This paper discusses about Clustering technique called Agglomerative hierarchical clustering using Tanagra tool. It groups patents with similar characteristics into one cluster based on between sum of square (BSS ratio) and Gap parameters.

The aim is to formulate a paradigm which is capable of reflecting the most innovative practices in delivering Tele-Medicare to patients, specifically focusing on rural and remote parts of a region. Tele health Therapy is not a practice area but a service delivery model that surpasses practice areas and improves access to occupational medical therapy services. The work presents an e-Medicare service (Tele Health Therapy) which is capable of providing immediate and appropriate treatment at the early stages of the medical emergencies. The urge of providing health services instantly, on demand, encourages researchers to formulate a system which can draw the health graph of the patient on the foundation of gathered parameters and compare it with ideal graph formulated on the basis of health form filled by the patient and with immediate relief graph. Tele Health Therapy attempts to deliver initial treatment with the help of diffusion of mobile technology, fuzzy database and ambient intelligence. Tele health Therapy diagnoses the person, in near real time and prescribes the initial level of medication comforting the patient in recovering to the stable state till the actual medical-aid can reach them. Tele-Health Therapy addresses the apprehension: "If there is anything the person is not sure about its health, then they should take the opportunity to ask any questions and raise any concerns they may have."

The growth in involvement of social media with current business world has influenced online customers by unbalanced opinions about the product and services. The main objective of this paper is to standardize the opinion given by the masses about any product or services in various social media communities. Enormous opinions on a specific product or service is available due to vast exposure available for publicly voice their opinion through social media. There are enormous customer reviews available about various products and services which are not systematically arranged. Moreover, there is no proper mechanism to identify the reliable or genuine reviews. This motivates the organizations and researchers to create tools which can automatically analyze and systematically arrange only those opinions that are genuine and filter out the fake reviews. The principal focus is to frame a filtering model with the help of truth test. Every source needs to undergo through a truth test, if the source is genuine, then the opinion is concluded while making a decision else it is marked as a fake review and hence filtered from the system.

This paper proposed a security compliance model that presents security and access control features which are employed at the time of origin of Big Data. The growth rate of data dealing devices and generation of data has increased exponentially which led to the massive explosion of data availability both in structured or unstructured form also known as big data. The challenge is to analyze this unstructured form of data and categories in a momentous form.The tool design to process big data without considering the security and access control feature in mind that its security is not inbuilt in these systems. Moreover, because of the huge size data it becomes a very tedious task to incorporate security feature at a later stage. So the security should be embedded in the initial or design stage.

Expeditious in silico determinations of the free energies of binding of a series of inhibitors to an enzyme are of immense practical value in structure-based drug design efforts. Some recent advances in the field of computational chemistry have rendered a rigorous thermodynamic treatment of biologic molecules feasible, starting from a molecular description of the biomolecule, solvent, and salt. Pursuing the goal of developing and making available a software for assessing binding affinities, we present here a computationally rapid, albeit elaborate, methodology to estimate and analyze the molecular thermodynamics of enzyme-inhibitor binding with crystal structures as the point of departure. The complexes of aspartic proteinases with seven inhibitors have been adopted for this study. The standard free energy of complexation is considered in terms of a thermodynamic cycle of six distinct steps decomposed into a total of 18 well-defined components. The model we employed involves explicit all-atom accounts of the energetics of electrostatic interactions, solvent screening effects, van der Waals components, and cavitation effects of solvation combined with a Debye-Huckel treatment of salt effects. The magnitudes and signs of the various components are estimated using the AMBER parm94 force field, generalized Born theory, and solvent accessibility measures. Estimates of translational and rotational entropy losses on complexation as well as corresponding changes in the vibrational and configurational entropy are also included. The calculated standard free energies of binding at this stage are within an order of magnitude of the observed inhibition constants and necessitate further improvements in the computational protocols to enable quantitative predictions. Some areas such as inclusion of structural adaptation effects, incorporation of site-dependent amino acid pKa shifts, consideration of the dynamics of the active site for fine-tuning the methodology are easily envisioned. The present series of studies, nonetheless, creates potentially useful qualitative information for design purposes on what factors favor protein-drug binding. The net binding free energies are a result of several competing contributions with 6 of the 18 terms favoring complexation. The nonelectrostatic contributions (i.e., the net van der Waals interactions) and the differential cavitation effects favor binding. Electrostatic contributions show considerable diversity and turn out to be favorable in a consensus view for the seven aspartic proteinase-inhibitor complexes examined here. Implications of these observations to drug design are discussed.