Preferred Language
Articles
/
bsj-9743
The Effect Of Optimizers On The Generalizability Additive Neural Attention For Customer Support Twitter Dataset In Chatbot Application

When optimizing the performance of neural network-based chatbots, determining the optimizer is one of the most important aspects. Optimizers primarily control the adjustment of model parameters such as weight and bias to minimize a loss function during training. Adaptive optimizers such as ADAM have become a standard choice and are widely used for their invariant parameter updates' magnitudes concerning gradient scale variations, but often pose generalization problems. Alternatively, Stochastic Gradient Descent (SGD) with Momentum and the extension of ADAM, the ADAMW, offers several advantages. This study aims to compare and examine the effects of these optimizers on the chatbot CST dataset. The effectiveness of each optimizer is evaluated based on its sparse-categorical loss during training and BLEU in the inference phase, utilizing a neural generative attention-based additive scoring function. Despite memory constraints that limited ADAMW to ten epochs, this optimizer showed promising results compared to configurations using early stopping techniques. SGD provided higher BLEU scores for generalization but was very time-consuming. The results highlight the importance of finding a balance between optimization performance and computational efficiency, positioning ADAMW as a promising alternative when training efficiency and generalization are primary concerns.

Scopus Crossref
View Publication Preview PDF
Quick Preview PDF
Publication Date
Tue Dec 01 2020
Journal Name
Baghdad Science Journal
A Modified Support Vector Machine Classifiers Using Stochastic Gradient Descent with Application to Leukemia Cancer Type Dataset

Support vector machines (SVMs) are supervised learning models that analyze data for classification or regression. For classification, SVM is widely used by selecting an optimal hyperplane that separates two classes. SVM has very good accuracy and extremally robust comparing with some other classification methods such as logistics linear regression, random forest, k-nearest neighbor and naïve model. However, working with large datasets can cause many problems such as time-consuming and inefficient results. In this paper, the SVM has been modified by using a stochastic Gradient descent process. The modified method, stochastic gradient descent SVM (SGD-SVM), checked by using two simulation datasets. Since the classification of different ca

... Show More
Scopus (10)
Crossref (6)
Scopus Clarivate Crossref
View Publication Preview PDF
Publication Date
Tue Dec 01 2020
Journal Name
Al-khwarizmi Engineering Journal
Applying Trade-off Curve to Support Set-Based Design application at an Aerospace Company

Companies compete greatly with each other today, so they need to focus on innovation to develop their products and make them competitive. Lean product development is the ideal way to develop product, foster innovation, maximize value, and reduce time. Set-Based Concurrent Engineering (SBCE) is an approved lean product improvement mechanism that builds on the creation of a number of alternative designs at the subsystem level. These designs are simultaneously improved and tested, and the weaker choices are removed gradually until the optimum solution is reached finally. SBCE implementations have been extensively performed in the automotive industry and there are a few case studies in the aerospace industry. This research describe the use o

... Show More
Crossref (1)
Crossref
View Publication Preview PDF
Publication Date
Thu Jun 20 2019
Journal Name
Baghdad Science Journal
An Analysis on the Applicability of Meta-Heuristic Searching Techniques for Automated Test Data Generation in Automatic Programming Assessment

Automatic Programming Assessment (APA) has been gaining lots of attention among researchers mainly to support automated grading and marking of students’ programming assignments or exercises systematically. APA is commonly identified as a method that can enhance accuracy, efficiency and consistency as well as providing instant feedback on students’ programming solutions. In achieving APA, test data generation process is very important so as to perform a dynamic testing on students’ assignment. In software testing field, many researches that focus on test data generation have demonstrated the successful of adoption of Meta-Heuristic Search Techniques (MHST) so as to enhance the procedure of deriving adequate test data for efficient t

... Show More
Crossref (2)
Clarivate Crossref
View Publication Preview PDF
Publication Date
Wed Jun 01 2022
Journal Name
Baghdad Science Journal
Variable Selection Using aModified Gibbs Sampler Algorithm with Application on Rock Strength Dataset

Variable selection is an essential and necessary task in the statistical modeling field. Several studies have triedto develop and standardize the process of variable selection, but it isdifficultto do so. The first question a researcher needs to ask himself/herself what are the most significant variables that should be used to describe a given dataset’s response. In thispaper, a new method for variable selection using Gibbs sampler techniqueshas beendeveloped.First, the model is defined, and the posterior distributions for all the parameters are derived.The new variable selection methodis tested usingfour simulation datasets. The new approachiscompared with some existingtechniques: Ordinary Least Squared (OLS), Least Absolute Shrinkage

... Show More
Scopus (3)
Crossref (1)
Scopus Clarivate Crossref
View Publication Preview PDF
Publication Date
Thu Dec 28 2023
Journal Name
Al–bahith Al–a'alami
Digital Interactive Elements in The Twitter Platform: An Analytical Study

The primary issue addressed in this research revolves around identifying the interactive elements provided by the Twitter platform and understanding their utilization by digital newspapers with official accounts. These newspapers broadcast content in line with their policies. This study is classified within descriptive research that employed a survey method and content analysis tool. The methodology relies on the "how was it said?" approach to categorize the analysis. The research yielded the following results:
Twitter utilized numerous interactive elements for disseminating tweets, which include "text, branching links, hashtags, digital images, digital videos, digital audio, and digital polls." However, thes

... Show More
Crossref
View Publication Preview PDF
Publication Date
Tue Dec 01 2020
Journal Name
Baghdad Science Journal
Detection of Suicidal Ideation on Twitter using Machine Learning & Ensemble Approaches

Suicidal ideation is one of the most severe mental health issues faced by people all over the world. There are various risk factors involved that can lead to suicide. The most common & critical risk factors among them are depression, anxiety, social isolation and hopelessness. Early detection of these risk factors can help in preventing or reducing the number of suicides. Online social networking platforms like Twitter, Redditt and Facebook are becoming a new way for the people to express themselves freely without worrying about social stigma. This paper presents a methodology and experimentation using social media as a tool to analyse the suicidal ideation in a better way, thus helping in preventing the chances of being the victim o

... Show More
Scopus (25)
Crossref (16)
Scopus Clarivate Crossref
View Publication Preview PDF
Publication Date
Thu Jun 01 2017
Journal Name
Journal Of Economics And Administrative Sciences
Measure the quality of Islamic banking customer perspective, Empirical Study on some Islamic banks Arabia

This study aims to identify the level of quality of  Islamic banking services in the Kingdom of Saudi Arabia. It also investigates the importance of  the various dimensions of the quality of banking services. The sample of the study included 69 clients of Al Rajhi Bank and  Al-Bilad Bank. The study concluded that the overall level quality of services that Islamic banking offer is high. However, their relative importance varies from one dimension to another. The security, confidence and physical aspects are very high which this is a good point. On the other hand, the application aspects do not get the same importance. These aspects include the customer's wishes, helping them to improve the services offered to them,

... Show More
Crossref (1)
Crossref
View Publication Preview PDF
Publication Date
Thu Sep 01 2011
Journal Name
Journal Of Economics And Administrative Sciences
Opportunities for establishing business incubators in IraqTo support microenterprises

Business incubator is a new effective mechanism in developing small projects through its introductions of a new integral system of services. It aims at supporting and developing making new projects. Hence, there is a big number of factors that are interrelated in the processes of preparation for those projects. Those factors are: organizing the incubator and the market available for the projects attached to them and the work programs which will have to be implemented. Those small projects represent more than 98% of the total work institutions in the world. Also it has become responsible for a ration reaching half of the national output of those countries. These projects have created between 40 to 80% of job opportunities availabl

... Show More
Crossref
View Publication Preview PDF
Publication Date
Sun Mar 01 2020
Journal Name
Baghdad Science Journal
An Application of Non-additive Measures and Corresponding Integrals in Tourism Management

Non-additive measures and corresponding integrals originally have been introduced by Choquet in 1953 (1) and independently defined by Sugeno in 1974 (2) in order to extend the classical measure by replacing the additivity property to non-additive property. An important feature of non –additive measures and fuzzy integrals is that they can represent the importance of individual information sources and interactions among them. There are many applications of non-additive measures and fuzzy integrals such as image processing, multi-criteria decision making, information fusion, classification, and pattern recognition. This paper presents a mathematical model for discussing an application of non-additive measures and corresp

... Show More
Scopus (12)
Crossref (5)
Scopus Clarivate Crossref
View Publication Preview PDF
Publication Date
Thu Aug 01 2019
Journal Name
Journal Of Economics And Administrative Sciences
Effect of the application of IFRS 15 "Revenue from contracts with customers" on the quality of financial reporting

Abstract

This study aims at identifying the impact of the application of IFRS 15 "Revenue from contracts with customers on the quality of financial reporting, through application to faculty members in the accounting departments of Iraqi universities and auditors. The problem of the study was the multiplicity of accounting rules and standards Which deals with the issues of revenue recognition , as well as the lack of consistency of most of them with the common framework of financial accounting, which results in low quality of financial reporting in the current financial statements, where the formulation of one hypothesis was the lack of relationship of significant significance The application of IFRS 15 "Recognition of rev

... Show More
Crossref
View Publication Preview PDF