Statistical methods

This paper investigates the impact of word embedding techniques on enhancing SMS spam detection models. Traditional statistical methods (BoW, TF-IDF) are compared with advanced techniques (Word2Vec, fastText, GloVe, PhoBERT) using a proprietary dataset.

5p viengfa 28-10-2024 2 2   Download

The paper used secondary data collected from the branch and primary data gathered from 50 representative customers and 10 credit officers related the group lending of the branch. Data analysis methods consisted of descriptive statistics and comparative analysis, incorporated with the measurements for bank performance and quality.

10p vibecca 01-10-2024 4 2   Download

This study was conducted to analyze water quality in a coastal province of Vietnam. Multivariate statistical methods, namely cluster analysis (CA) and principal component analysis (PCA) were utilized.

12p vibecca 01-10-2024 2 1   Download

This paper aimed to investigate the actors involved in the orange supply chain in the province. The study used descriptive statistics and chain analysis methods to explore the problems in each stage of the orange supply chain.

13p vibecca 01-10-2024 3 2   Download

This study, therefore, investigated the determinants of the labor productivity of garment firms in Nam Dinh province. The data used for this study were extracted from the Enterprise Survey Dataset in 2021, and descriptive statistics, comparative analysis, and the Cobb-Douglas function in logarithm form were the major methods employed for the study.

9p vibecca 01-10-2024 4 2   Download

The study which applies quantitative methods along with the support of EVIEWS software points out a statistical significance relationship. Firstly, the CS has a contrariwise impact on ROA but no effect on ROE and ROS. Secondly, among control variables (asset structure, expenditure ratio, revenue), the expenditure ratio is the only one showing a contrariwise impact on ROS and the rest has no statistical significance.

10p viormkorn 27-09-2024 7 2   Download

Several conventional research methods such as data collection, analysis, statistics, comparison, studies of historical experience and forecasting for a reconsideration of the values and usability of bamboo in traditional housing and modern architecture are introduced in the paper in order to accomplish the objectives to be set.

7p viyoko 24-09-2024 3 1   Download

Distribution of the proportion of bacterial isolates and their antibiotic resistance rate at Ninh Thuan Provincial General Hospital (NTPGH). Subjects and methods: All bacterial isolates from various types of clinical specimens from a departments of NTPGH from October 1, 2022 to September 30, 2023.

9p vifaye 20-09-2024 11 1   Download

This paper is conducted to investigate the relationship between customer satisfaction and firm performance within the businesses that are the retailers of technological devices such as smartphones, laptops, desktop computers, and peripheral devices in Vietnam. The research method used in this study is a quantitative method with multivariate statistical analysis steps such as testing by Cronbach's Alpha coefficient, explore descriptive analysis, correlation analysis, exploratory factor analysis, and multi-stepwise regression analysis.

18p toduongg 24-08-2024 0 0   Download

The purpose of our study was to evaluate the effectiveness of this method in treating femoral shaft nonunion after open reduction and internal fixation with locked intramedullary nail fixation.

6p vinatisu 29-08-2024 1 0   Download

The article focuses on analyzing and evaluating a number of situations related to the substainable production and export of agricultural products of Vietnam in the context that trade protection measures continue to increase in the current period. Using qualitative research methods such as statistics, description, comparison, contrast, synthesis and analysis, ..., the research results show that trade protection measures, especially the Non-tariff measures have been and are tending to be more diversified and more sophisticated.

15p toduongg 24-08-2024 2 2   Download

This research study investigates the motivation for English language learning among undergraduate students at the University of Transport and Communications (UTC). The study adopts a mixed-methods approach, combining qualitative and quantitative methods to gather data. It includes a literature review on motivation theories, language learning, and questionnaire surveys. Content analysis is used for qualitative data, while descriptive and inferential statistics are used for quantitative data analysis.

7p tuongbachxuyen 05-08-2024 3 1   Download

The article discloses methodological approaches to quantifying the level of penetration of block chain technologies into the regional socioeconomic environment in comparison to the supply chain strategies. Based on the methods of mathematical and statistical analysis, it allows you to formalize and convert a large array of data of a qualitative order into a single scale for measuring the process under study - the level of demand in the regional socio-economic environment for block chain technology.

6p longtimenosee09 08-04-2024 1 1   Download

This aim of this study is to measure the technical efficiency score and identify the factors that affect the technical efficiency in plastic manufacturing firm in Malaysia for the year 2015 using the two stage Data Envelopment Analysis (DEA) method. The first stage involves calculating the efficiency score through the DEA using firm-level data, provided by the Department of Statistics, Malaysia. In the second stage, Tobit Regression Analysis was used to identify the significant factors affecting the efficiency of the plastic industry.

9p longtimenosee09 08-04-2024 7 1   Download

One of the limitations of the BDI (Belief-Desire-Intention) model is the lack of any explicit mechanisms within the architecture to be able to learn. In particular, BDI agents do not possess the ability to adapt based on past experience. This is important in dynamic environments as they can change, causing previously successful methods for achieving goals to become inefficient or ineffective.

108p runthenight07 01-03-2023 10 3   Download

The project's main objectives were to analyse the data obtained during geological exploration and use it to model and predict biological patterns in order to answer the main research questions: 1) Is there a statistical relationship between the environmental parameters and the benthic community composition and morphospecies observed? 2) Can predictive models be developed from this relationship? In achieving these main objectives, the project also developed recommendations for methods that could be applied during geological exploration voyages to maximise the potential use of data collected.

206p runthenight04 02-02-2023 5 1   Download

This study intends to contribute to the current state of knowledge by empirically investigating the relationship between institutional environment and the performance of MFIs, by comparing South Asia and Latin America.

164p runthenight04 02-02-2023 4 3   Download

In present paper "Equation of state and thermal expansion of metals with FCC structure: Application to Cu, Al and Ni", by the statistical moment method (SMM) we obtain the equation of state for metals with cubic structure at temperature 0 K and pressure P and from that obtain the expressions of lattice parameter and thermal expansion coefficient for these metals at temperature T and pressure P.

6p runordie3 27-06-2022 13 2   Download

Compared to RNA-sequencing transcript differential analysis, gene-level differential expression analysis is more robust and experimentally actionable. However, the use of gene counts for statistical analysis can mask transcript-level dynamics. We demonstrate that ‘analysis first, aggregation second,’ where the p values derived from transcript analysis are aggregated to obtain gene-level results, increase sensitivity and accuracy.

11p vimichaelfaraday 25-03-2022 13 3   Download

This project proposes two new methods of deep neural networks and handcrafted features for damage detection. The first method uses a convolution neural network (CNN) to extract deep features in time series and Long Short Term Memory (LSTM) network to find a statistically significant correlation of each lagged feature in time series data.

6p billyelliot 11-11-2021 28 1   Download