Business Data Analytics Applications to Online Product Reviews and Nationalism

This paper investigates the data analytics between consumer purchase decisions relative to the on-line reviews. The multi-attributes associated with purchase decisions are comprised of nationalism and consumer preference to be correlated with online reviews using big data analytics. By far, a small fraction of meaningful studies have sought to correlate nationalism and ethnocentrism with big data analytics to date. Globally accepted generic products are selected to expedite the process of data engineering. Two sets were arranged: passenger automobiles for transportation with an estimated $9 trillion global market and the smart phone, boosting its market size of approximately $5 billion. Both products provide minimized cultural, linguistic, gender, age, and/or custom barriers of entry for prospective digital consumers, thereby allowing relatively unrestricted engagement with online reviews and purchases. A series of hypothesis tests indicate that there is a positive correlation between nationalism and automobiles. As to smart cell phones, however, nationalism had nominal control factors. Multi-variate analytics were performed by using R and Tableau Public.

COVID-19 Sentiments and Impact on Stock Market Prices

This paper studies the impact of sentiments expressed by tweets from Twitter on the stock market associated with COVID-19 during the critical period from December 1, 2019 to May 31, 2020. The stock prices of 30 companies on the Dow Jones Index were collected for this period. Twitter tweets were also collected, using the search phrases “COVID-19” and “Corona Virus” for the same period, and their sentiment scores were calculated. The three time series, open and close stock values, and the corresponding sentiment scores from tweets were sorted by date and combined. Multivariate time series models based on vector error correction (VEC) models were applied to this data. Forecasts for these 30 companies were made for the time series open, for the 30 days of June 2020, following the data collection period. Stock market data for the month of June was for all the companies was compared with the forecast from the model. These were found to be in excellent agreement, implying that sentiment had a significant impact or was significantly impacted by the stock market prices.

An Experimental Data of Lithium-Ion Battery Time Series Analysis

The experimental data of Lithium-ion battery has its specific sense. This paper is proposed to analyze and forecast it by using autoregressive integrated moving average (ARIMA) and spectral analysis, which has effective and statistical results. The method includes the identification of the data, estimation and diagnostic checking, and forecasting the future values by Box and Jenkins. The analysis shows that the time series models are related with the present value of a series to past values and past prediction errors. After transferring the data by different function, improving autocorrelations are significant. Forecasting the future values of the possible observations show significantly fluctuated such as increasing or decreasing in specific ranges accordingly. In spectral analysis, the parameters of the model were determined by performing spectral analysis of the experimental data to look periodicities or cyclical patterns, and to check the existence of white noise in the data. The Bartlett's Kolmogorov-Smirnov statistic suggests the white noise of the data. The spectral analysis for the series reveals non-11-second cycle of activity for dynamic stress test current, but strong 45-second that highlights the position of the main peak in the spectral density; strong 21-second and 45-second for the urbane dynamometer driver schedule current and voltage, respectively; but no significance for dynamic stress test current.

Fitting a Three-Phase Discrete SIR Model to New Coronavirus Cases in New York State

CDC data on new coronavirus cases in New York State between March 4, 2020 and June 26, 2020 show three distinct phases for the spread of the virus. The authors demonstrate fitting of a simple discrete SIR model with three phases to model these data, achieving a high fidelity to the data. Optimal model fits using both R and Excel are compared, and various issues are discussed. Finally, the model for New York State is treated as a training set for extending and applying the model to the outbreak in other areas of the United States and the country as a whole.

Social Media Content Analysis and Classification Using Data Mining and ML

Students' natural conversations on social media such as Twitter and Whatsapp are useful to understand their learning experiences feelings. Collecting and analyzing data from such media can be a difficult task. However, the large scale of data is required for automatic data analysis techniques to classify Twitter data. The proposed new system is a combination of qualitative analysis and large-scale data mining and ML techniques. This system focuses on engineering students' Twitter posts, which are collected from engineering colleges, to understand issues and problems in their learning. The authors first conduct a qualitative analysis using ML studio on tweets collected from engineering colleges using term #DStudentsproblems, engineeringProblem, Aluminisuggestions, and ladyEngineer. Collected tweets are related to engineering students' college lives. In the proposed system, a multi-label classification algorithm to classify tweets reflecting students' problems such as soft skill issues, heavy study load, lack of social engagement, and sleep problems is used.

Characterization and Predictive Analysis of Volatile Financial Markets Using Detrended Fluctuation Analysis, Wavelet Decomposition, and Machine Learning

This paper proposes a granular framework for examining the dynamics of stock indexes that exhibit nonparametric and highly volatile behavior, and subsequently carries out the predictive analytics task by integrating detrended fluctuation analysis (DFA), maximal overlap discrete wavelet transformation (MODWT), and machine learning algorithms. DFA test ascertains the key temporal characteristics of the daily closing prices. MODWT decomposes the time series into granular components. Four pattern recognition algorithms—adaptive neuro fuzzy inference system (ANFIS), dynamic evolving neural-fuzzy inference system (DENFIS), bagging and deep belief network (DBN)—are then used on the decomposed components to obtain granular level forecasts. The entire exercise is performed on daily closing prices of Dow Jones Industrial Average (DJIA), National Stock Exchange of India (NIFTY), Karachi Stock Exchange (KSE), Taiwan Stock Exchange (TWSE), Financial Times Stock Exchange (FTSE), and German Stock Exchange (DAX). MODWT-Bagging and MODWT-DBN appear as superior forecasting models.

Data Analysis and Visualization in Python for Polar Meteorological Data

The aim of this study is to analyze meteorological data obtained from the various expeditions made to the Indian stations in Antarctica over recent years and determine how significantly the weather has shown a marked change over the years. For any time series data analysis, there are two main goals: (a) the authors need to identify the nature of the phenomenon from the sequence of observations and (b) predict the future data. On account of these goals, the pattern in the time series data and its variability are to be accurately identified. This paper can then interpret and integrate the pattern established with its associated meteorological datasets collected in Antarctica. Using the data analytics knowledge the validity of interpretation for the given datasets a pattern has been identified, which could extrapolate the pattern towards prediction. To ease the time series data analysis, the authors developed online meteorological data analytic portal at NCPOR, Goa http://data.ncaor.gov.in/.

View Materialization Over Big Data

Advances in technology have resulted in the generation of a large volume of heterogeneous big data for large enterprises engaged in e-commerce, healthcare, education, etc. This is being created at a rapid rate but is low in its veracity. This big data includes large sets of semi-structured and unstructured data and is stored over a distributed file system (DFS). This data can be processed in a fault tolerant manner using several frameworks, tools, and advanced database technologies. Big data can provide important information, which can be used for business decision making. View materialization, which has been widely studied for structured databases or data warehouse, has been extended to big data to enhance efficiency of big data query processing. This paper focuses on the selection of big data views for materialization. The big data views can be identified by extracting a set of query attributes from the set of query workload of an enterprise. The query attributes are interrelated resulting in the creation of alternate access paths for query evaluation. The cost of query processing using big data views involves the integrity of different data types of heterogeneous big data, frequency of queries, change in the size of big data, selected sets of big data materialized views, and updates on big data and these sets of materialized views. The cost of query processing is computed using the stored size of big data views on the DFS system, which is a consistent processing framework of DFS. A big data view selection algorithm that is capable of selecting views from structured, semi-structured, and unstructured data has been proposed in this paper. The proposed algorithm would select big data views that would result in faster processing of most user queries resulting in efficient decision making.

Perspectives of Fuzzy Logic and Their Applications

Fuzzy logic is a highly suitable and applicable basis for developing knowledge-based systems in engineering and applied sciences. The concepts of a fuzzy number plays a fundamental role in formulating quantitative fuzzy variable. These are variable whose states are fuzzy numbers. When in addition, the fuzzy numbers represent linguistic concepts, such as very small, small, medium, and so on, as interpreted in a particular contest, the resulting constructs are usually called linguistic variables. Each linguistic variable the states of which are expressed by linguistic terms interpreted as specific fuzzy numbers is defined in terms of a base variable, the value of which are real numbers within a specific range. A base variable is variable in the classical sense, exemplified by the physical variable (e.g., temperature, pressure, speed, voltage, humidity, etc.) as well as any other numerical variable (e.g., age, interest rate, performance, salary, blood count, probability, reliability, etc.). Logic is the science of reasoning. Symbolic or mathematical logic is a powerful computational paradigm. Just as crisp sets survive on a 2-state membership (0/1) and fuzzy sets on a multistage membership [0 - 1], crisp logic is built on a 2-state truth-value (true or false) and fuzzy logic on a multistage truth-value (true, false, very true, partly false and so on). The author now briefly discusses the crisp logic and fuzzy logic. The aim of this paper is to explain the concept of classical logic, fuzzy logic, fuzzy connectives, fuzzy inference, fuzzy predicate, modifier inference from conditional fuzzy propositions, generalized modus ponens, generalization of hypothetical syllogism, conditional, and qualified propositions. Suitable examples are given to understand the topics in brief.

Fuzzy Bayesian Context-Aware System to Reduce Electricity Consumption

Context-awareness develops smart, intelligent IoT devices that can adapt to changing needs and act autonomously on behalf of the user. The main challenge of context-aware internet of things is to interpret the context effectively. There is an abundance of CAIOT in literature. Understanding of the meaning of the context is, however, almost ignored. Misinterpretation of context can lead to an incorrect decision that motivates to develop a system that emphasis context reasoning and decision making using the fuzzy Bayesian approach. The current investigation aims to build a context-aware IoT system using occupancy detection for energy management. The performance evaluation for the proposed system uses data collected in the tutorial room to detect occupancy. Extensive experiments highlight the utility of the proposed approach, which significantly reduces energy than the traditional ON/OFF usage pattern through customer access via mobile phone or personal computer.