Forecasting the efficiency of weft knitting production: a decision tree method

Purpose This paper aims to develop a decision support system for predicting the knitting production’s efficiency based on the input parameters of an order. This tool supports the operations managers to make reliable decisions of estimated delivery time, which will result in reducing waste arising from late delivery, overtime and increased labor. Design/methodology/approach The decision tree method with a set of logical IF-THEN rules is used to determine the knitting production’s efficiency. Each path of the decision tree represents a rule of the following form: “IF <Condition> THEN <Efficiency label>.” Starting with identifying and categorizing input specifications, the model is then applied to the observed data to regenerate the results of efficiency into classification instances. Findings The production’s efficiency is the result of the interaction between input specifications such as yarn’s component, knitting fabric specifications and machine speed. The rule base is generated through a decision tree built to classify the efficiency into five levels, including very low, low, medium, high and very high. Based on this, production managers can determine the delivery time and schedule the manufacturing planning more accurately. In this research, the correct classification instances, which is simply a ratio of the correctly predicted observations to the total ones, reach 80.17%. Originality/Values This research proposes a new methodology for estimating the efficiency of weft knitting production based on a decision tree method with an application of real data. This model supports the decision-making process of the estimated delivery time.

Perbandingan Metode KNN, Decision Tree, dan Naïve Bayes Terhadap Analisis Sentimen Pengguna Layanan BPJS

BPJS is really helpful because one of its goal is to provide good service for the member in terms of healthiness. But, when there’s many people using the service, then it will cause more pros and contras. Therefore, researcher will be doing sentiment analysis in the field of data mining towards bpjs users on social media Twitter as much as 1000 data that later will be filtered to be 903 data because there are some data that has been duplicated. Researchers used the KNN, Decision Tree, and Naïve Bayes methods to compare the accuracy of the three methods. Researchers used the RapidMiner version 9.7.2 tools. The results showed that the sentiment analysis of Twitter data on BPJS services using the KNN method reached an accuracy level of 95.58% with class precision for pred. negative is 45.00%, pred. positive is 0.00%, and pred. neutral is 96.83%. Then the Decision Tree method the accuracy rate reaches 96.13% with the precision class for pred. negative is 55.00%, pred. positive is 0.00%, and pred. neutral is 97.28%. And the last one is the Naïve Bayes method which achieves 89.14% accuracy with precision class for pred. negative is 16.67%, pred. positive was 1.64%, and pred. neutral is 98.40%.

The reasons why the Regression Tree Method is more suitable than General Linear Model to analyze complex educational datasets

Any quantitative method is shaped by certain rules or assumptions which constitute its own rationale. It is not by chance that these assumptions determine the conditions and constraints which permit the evidence to be constructed. In this article, we argue why the Regression Tree Method’s rationale is more suitable than General Linear Model to analyze complex educational datasets. Furthermore, we apply the CART algorithm of Regression Tree Method and the Multiple Linear Regression in a model with 53 predictors, taking as outcome the students’ scores in reading of the 2011’s edition of the National Exam of Upper Secondary Education (ENEM; N = 3,670,089), which is a complex educational dataset. This empirical comparison illustrates how the Regression Tree Method is better suitable than General Linear Model for furnishing evidence about non-linear relationships, as well as, to deal with nominal variables with many categories and ordinal variables. We conclude that the Regression Tree Method constructs better evidence about the relationships between the predictors and the outcome in complex datasets.

Hardware Reliability Analysis of a Coal Mine Gas Monitoring System Based on Fuzzy-FTA

The hardware reliability of a gas monitoring system was investigated using the fuzzy fault tree analysis method. A fault tree was developed considering the hardware failure of the gas monitoring system as a top event. Two minimum path sets were achieved through qualitative analysis using the ascending method. The concept of fuzzy number of the fuzzy set theory was applied to describe the probability of basic event occurrence in the fault tree, and the fuzzy failure probabilities of the middle and top events were calculated using fuzzy AND and OR operators. The results show that the proposed fuzzy fault tree is an effective method of reliability analysis for gas monitoring systems. Results of calculations using this method are more reasonable than those obtained with the conventional fault tree method.

Disclosure Sentiment: Machine Learning vs. Dictionary Methods

We compare the ability of dictionary-based and machine-learning methods to capture disclosure sentiment at 10-K filing and conference-call dates. Like Loughran and McDonald [Loughran T, McDonald B (2011) When is a liability not a liability? Textual analysis, dictionaries, and 10-Ks. J. Finance 66(1):35–65.], we use returns to assess sentiment. We find that measures based on machine learning offer a significant improvement in explanatory power over dictionary-based measures. Specifically, machine-learning measures explain returns at 10-K filing dates, whereas measures based on the Loughran and McDonald dictionary only explain returns at 10-K filing dates during the time period of their study. Moreover, at conference-call dates, machine-learning methods offer an improvement over the Loughran and McDonald dictionary method of a greater magnitude than the improvement of the Loughran and McDonald dictionary over the Harvard Psychosociological Dictionary. We further find that the random-forest-regression-tree method better captures disclosure sentiment than alternative algorithms, simplifying the application of the machine-learning approach. Overall, our results suggest that machine-learning methods offer an easily implementable, more powerful, and reliable measure of disclosure sentiment than dictionary-based methods. This paper was accepted by Brian Bushee, management science.