Movie Success Prediction Using Data Mining

The movie industry is arguably one of the biggest entertainment sectors. Nollywood, the Nigerian movie industry produces tons of movies for public consumption, but only a few make it to box-office or end up becoming blockbusters. The introduction of movie success prediction can play an important role in the industry not only to predict movie success but to help directors and producers make better decisions for the purpose of profit. This study proposes a movie prediction model that applies data mining techniques and machine learning algorithms to predict the success or failure of an upcoming movie (based on predefined parameters). The parameters needed for predicting the success or failure of a movie include dataset needed for the process of data mining such as the historical data of actors, actresses, writers, directors, marketing and production budget, audience, location, release date, and competing movies on same release date. This model also helps movie consumers to determine a blockbuster, hit, success rating and quality of upcoming movies before deciding on a movie ticket. The data mining techniques was applied to Internet Movie Database MetaData which was initially passed through cleaning and integration process.

Automated Teller Machine-Based Voting System

Voting is a critical element of any election which involves the processes of electing leaders or representatives into positions of authority in a democratic system of government. In most developing countries of the world, this process is usually marred with challenges of confidentiality, integrity, availability and auditability such as falsification of results, identity theft, theft of ballot boxes, multiple voting problems, over voting, and electoral fraud. This paper presents a framework for Automated Teller Machine-based voting system that solves the aforementioned challenges of the current voting system by using the existing Automated Teller Machines and debit cards issued for voting. Going further to implement the solution proposed in this paper will enhance and guarantee the credibility of the electoral processes and show a true reflection of the wishes of the people.

Predicting Foreign Exchange Using Digital Signal Processing

The forex market is one associated with so much volatility and can lead to grave financial losses if not properly understood. To understand the market is to study the price patterns from previous years or months and make predictions from the rate of falling and rising. There have been so much researches aimed at developing a predictive model for the FOREX market, however, no model has been able to handle the market volatility while predicting future rates accurately. In this work, we have developed a digital processing model for predicting foreign exchange using ARIMA and Artificial Neural Network algorithms. We used price datasets for five currencies namely: USD, Swiss Pounds, Yen, Euro and Franc, gotten from the Central Bank of Nigeria (CBN) website. The data ranged from a period of 20 years. The model was simulated using MATLAB software. The study performed excellently in terms of time (26 seconds) and minimal errors (0.7). This work could be beneficial to FOREX traders and to the entire research community.

Digital Signal Processing for Predicting Stock Prices

With the exponential growth of big data and data warehousing, the amount of data collected from various stock markets around the world has increased significantly. It is now impossible to process and analyze data using mathematical techniques and basic statistical calculations to forecast trends such as closing and opening prices, as well as daily stock market lows and highs. The development of smart and automated stock market forecasting systems has made significant progress in recent years. Digital signal processing is required for analysis and preprocessing because of the accuracy and speed with which these large amounts of data must be processed and analyzed. In this paper, we evaluate some of these predictive algorithms based on three parameters such as speed, accuracy and complexity, we analyze the data using the dataset from kaggle.com and we implement these algorithms using pythons. The results of our analysis in this paper shows a significant correlation between the yearly prices until the year 2018 where there is a significant increase in stock price.

Implementation of Greedy Algorithm in Optimal Cellular Masts Hoisting Over Population Coverage

The need to meet the demands of subscribers of wireless services is imperative to Global System for Mobile (GSM) Communication companies. These demands, which revolve around the maintenance of good network coverage and improved Quality of Service (QoS), depend largely on the nature of cellular network masts. Greedy Algorithm Model was implemented on network masts for small size populated areas to effectively have optimal network coverage. Object Oriented Analysis and Design Methodology (OOADM) and Java programming language was used for its implementation. The analysis of the results shows that Greedy Algorithm performed optimizing cellular network masts hoisting optimally for small populated areas.

A Comparative Analysis of Image Compression using PCM and DPCM

The quality of the signal is essential in digital communication and signal processing. The transmission channel is also important. Modulation is used for effectively transmission of signal. There exist several types of modulation techniques. One of such is the pulse code modulation (PCM). The performance of PCM is however affected by quantization error and noise in the transmission channel, which affects the quality of the output. Against this backdrop, this paper presents the use of differential pulse code modulation (DPCM) so as to address the limitation of pulse code modulation. The simulation environment is MATLAB 2018a. The MATLAB Simulink is used to design the PCM and DPCM systems using appropriate digital processing blocks. The DPCM system shows a significant improvement in terms of error reduction and quality of output.

New K-means Clustering Method Using Minkowski’s Distance as its Metric

Cluster analysis is an unsupervised learning method that classifies data points, usually multidimensional into groups (called clusters) such that members of one cluster are more similar (in some sense) to each other than those in other clusters. In this paper, we propose a new k-means clustering method that uses Minkowski’s distance as its metric in a normed vector space which is the generalization of both the Euclidean distance and the Manhattan distance. The k-means clustering methods discussed in this paper are Forgy’s method, Lloyd’s method, MacQueen’s method, Hartigan and Wong’s method, Likas’ method and Faber’s method which uses the usual Euclidean distance. It was observed that the new k-means clustering method performed favourably in comparison with the existing methods in terms of minimization of the total intra-cluster variance using simulated data and real-life data sets.

Media, Information and Communication Technology and Sustainable Development in Nigeria

This study/paper/research work shows the role of Media, Information and Communication Technology (ICT) and sustainable development in Nigeria; it aimed to improve lives in many other ways, such as through education, skill development, new services creation, innovation and automation, freedom of expression through mass media and expose corrupt practices in Nigeria, although the primary responsibilities of the media is to entertain without words falsely spoken that damage the reputation of one another, inform based on accurate facts and educate on current relevant issues. But not withstanding Media, Information and Communication Technology make a significant impact in sustainable and development in Nigeria.

Deconstructing the Ethernet with ChokyFop

The analysis of digital-to-analog converters is an unproven challenge. After years of confirmed research into flip-flop gates, we prove the analysis of systems, which embodies the compelling principles of programming languages. In our research we propose a novel algorithm for the refinement of multi-processors (ChokyFop), confirming that information re- trieval systems can be made relational, wireless, and constant- time.

Analytical Models for the Computation of Error Probability of Multi-Level Phase Shift Keying Modulation Scheme

In this paper, analytical models for the computation of error probability (BER) of the Multi-level Phase Shift Keying (MPSK) modulation scheme is presented. Analytical models for computing MPSK bit error probability based on Q function, error function (erf) and complementary error function (erfc) are presented. Also, an analytical model for computing the symbol error rate for MPSK is presented. Furthermore, a generalized analytical expression for BER as a function of modulation order (M) and energy per bit to noise power density ratio (Eb/No) is presented. The BER was computed for various values of M (2 ≤ M ≤ 256) and Eb/No (0 dB ≤ Eb/No ≤ 14 Db). The results showed that at Eb/No =12 dB, a BER of 9.006E-09 is realized for M =2 and M =4 whereas BER of 1.056E-01 is realized for M = 256. Also, for the same M = 2 , the value of BER decreased from 1.2501E-02 at Eb/No = 4 dB to 9.0060E-09at Eb/No =12 dB. Generally, the results showed that for the MPSK modulation scheme, for a given value of Eb/No, the lower modulation order (M) has a lower BER and for a given modulation order, (M) the BER decreases as Eb/No increases.