beeqb,mx,github

shop_revenue_prediction

shops revenue

simple-linear-regression-to-determining-salaries

How exactly do companies, especially those mass recruiting multinational corporations determine salaries of thousands of new employees in one go? Does years of experience of employees really help them determine salaries on such a large scale? To answer the question, I tried to explore a sample salary dataset of an undisclosed company. It is interesting to know that using Simple Linear Regression (a basic Machine Learning algorithm), mass recruiters are deciding on salaries of thousands of employees in a matter of minutes, thereby cutting down on time spent, human resource cost and saving tens of thousands of dollars on a single mass recruitment drive. Dive into my github to read more. Approach I considered duration of work experience as x (independent variable) and salary as y (dependent variable). I then divided the dataset into training and test sets. I allotted random 80% of the observations to the training set (an act of making the machine learn) and based on the learning (training set) by machine, I tried to predict the salaries of the rest 20% of the employees. I then compared the predicted results with the actual test results of the dataset. I was intrigued to know that the predicted results were pretty close to the actual test results. This testifies how companies can and (some) are increasingly using machine learning algorithms in their recruitment processes. It is interesting how, using data, we can use a simple high school math concept to teach a machine and solve some of the fundamental business problems saving on time and money and drastically improving efficiency. In case there are multiple factors (independent variables) like work experience, qualification type, location of work etc that influence the salary (dependent variable), then multiple linear regression algorithm can be used.

soroka

Узнай, хорошо или плохо говорят о тебе или твоей фирме в Интернете! Наша "Сорока" с искусственным интеллектом принесёт тебе это на своём хвосте.

spendu

A finance management system that relies on the Autoregressive Integrated Moving Average Model to predict spending behaviour

statistical-arbitrage-model

For a family of stocks, generally belonging to the same sector or industry, there exists a correlation between prices of each of the stocks. There, though, exist anomalous times when for a small period of time, the correlation is broken. But the market self corrects in some time and the correlation is re-established. During this small window of time when correlation is anomalous, there exists a money-making opportunity for quantitative traders. Problem Statement: Develop Machine Learning Algorithm to predict statistical arbitrage opportunities in NSE based on the 2016 data. Test this algorithm on 2017 data.

stock-price-prediction

A group project for CMPE272 at SJSU

stockpriceprediction

Stock Price Prediction using Machine Learning Techniques

support-resistance

Visual style support and resistance detection using Python code

tensorflow-js-instagram

Detect fake instagram profiles with logistic regression and Tensorflow.js

tfaas

TensorFlow as a Service, a general purpose framework to serve TF models.

time-series-forecasting

Time-Series-Forecasting Time series analysis is a really popular area with a great number of applications in business, economics and finance and computer science. The time series data is used to provide visual information to the unpredictable nature of the market. Analysing the history, using periodic historical data to make decisions and plans on long-term forecasting. The assumptions are that the past pattern will continue in the future. In today’s organizations, all requirements of the business sector need accurate and practical reading into the future. Time series forecasting is becoming really crucial because it is a sign of the survival of business in the world. The forecast is an assumption of future values of the target variable. The forecast plays a significant role in finding the relationships between the audience and the TV series. The time series is a consecutive set of data points, measured typically over successive times. In mathematical form time series forecasting is a set of vectors x(t),t = 0,1,2,... where t represents the time elapsed. The variable x (t) is treated as a random variable. The measurements taken during time series are arranged in proper order. There are different types of time series data. If a time series contains a single variable it is called univariate. But if records of more than one variable are considered; it is termed as multivariate. Time series can be known as continuous if the observations are measured at every point of time, whereas a discrete time series has an observation measured at every instance of time. In a discrete time series the consecutive observations are recorded at equally spaced time intervals such as hourly, daily, weekly, monthly or yearly time separations. The potent aim of the time series analysis is to study the historical path of the data and to build a model to analyse data and to predict the future values of the time series. Time series forecasting is to build a model with the best possible accuracy and effectiveness. In this project sales data set is used for forecasting furniture sales.

traveladvisor

This Telegram bot helps you to choose a destination for your vacation. It predicts an average cost of the trip (tickets and the hotel) based on your preferences.