A Novel Method to Detect Segmentation points of Arabic Words using Peaks and Neural Network

Ramdan, Jabril; Omar, Khairuddin; Faidzul, Mohammad

doi:10.18517/ijaseit.7.2.1824

Cited by 9 publications

(7 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The pre-processing phase is applied to reduce noise and aberrations that may occur in acquired texts because of the limitations of the hardware or software used while writing the text. The noise or aberrations include irregular text size, non-centered text, missing some points of the lines and curves of the texts paths because of the high speeds of writing and uneven distances of points from neighbouring positions [11]. In this phase, several steps are included.…”

Section: Pre-processing Phasementioning

confidence: 99%

Online handwriting Arabic recognition system using k-nearest neighbors classifier and DCT features

Abuzaraida

Elmehrek

Elsomadi

2021

IJECE

View full text Add to dashboard Cite

With advances in machine learning techniques, handwriting recognition systems have gained a great deal of importance. Lately, the increasing popularity of handheld computers, digital notebooks, and smartphones give the field of online handwriting recognition more interest. In this paper, we propose an enhanced method for the recognition of Arabic handwriting words using a directions-based segmentation technique and discrete cosine transform (DCT) coefficients as structural features. The main contribution of this research was combining a total of 18 structural features which were extracted by DCT coefficients and using the k-nearest neighbors (KNN) classifier to classify the segmented characters based on the extracted features. A dataset is used to validate the proposed method consisting of 2500 words in total. The obtained average 99.10% accuracy in recognition of handwritten characters shows that the proposed approach, through its multiple phases, is efficient in separating, distinguishing, and classifying Arabic handwritten characters using the KNN classifier. The availability of an online dataset of Arabic handwriting words is the main issue in this field. However, the dataset used will be available for research via the website.

show abstract

Section: Pre-processing Phasementioning

confidence: 99%

Online handwriting Arabic recognition system using k-nearest neighbors classifier and DCT features

Abuzaraida

Elmehrek

Elsomadi

2021

IJECE

View full text Add to dashboard Cite

show abstract

“…2) Feature Scaling: Feature scaling or data normalization can be useful since the Occupancy data set has varying scales and the distribution is not Gaussian, that is a bell curve ([20]- [25]). It is particularly useful for classification algorithms such as k-Nearest Neighbours and Artificial Neural Networks [26][27][28]. The attributes in this data set is normalized by choosing the Unsupervised Attribute Normalize filter, in which each numeric attribute is rescaled to the range of 0 to 1.…”

Section: ) Descriptive Data Summarization and Data Visualization In mentioning

confidence: 99%

“…Table 7 shows that the occupancy minority class values have increased. [27]. This can be monitored and modified during training time.…”

Section: ) Descriptive Data Summarization and Data Visualization In mentioning

confidence: 99%

Redefining Selection of Features and Classification Algorithms for Room Occupancy Detection

Sani¹,

Shamsuddin²,

Sahran³

et al. 2018

Int. J. Adv. Sci. Eng. Inf. Technol.

View full text Add to dashboard Cite

The exponential growth of todays technologies has resulted in the growth of high-throughput data with respect to both dimensionality and sample size. Therefore, efficient and effective supervision of these data becomes increasing challenging and machine learning techniques were developed with regards to knowledge discovery and recognizing patterns from these data. This paper presents machine learning tool for preprocessing tasks and a comparative study of different classification techniques in which a machine learning tasks have been employed in an experimental set up using a data set archived from the UCI Machine Learning Repository website. The objective of this paper is to analyse the impact of refined feature selection on different classification algorithms to improve the prediction of classification accuracy for room occupancy. Subsets of the original features constructed by filter or information gain and wrapper techniques are compared in terms of the classification performance achieved with selected machine learning algorithms. Three feature selection algorithms are tested, specifically the Information Gain Attribute Evaluation (IGAE), Correlation Attribute Evaluation (CAE) and Wrapper Subset Evaluation (WSE) algorithms. Following a refined feature selection stage, three machine learning algorithms are then compared, consisting the Multi-Layer Perceptron (MLP), Logistic Model Trees (LMT) and Instance Based k (IBk). Based on the feature analysis, the WSE was found to be optimal in identifying relevant features. The application of feature selection is certainly intended to obtain a higher accuracy performance. The experimental results also demonstrate the effectiveness of Instance Based k compared to other ML classifiers in providing the highest performance rate of room occupancy prediction.

show abstract

“…ANN is used in widespread area of applications including the field of data mining, finance, process control, flight security, medical, marketing, pattern recognition, forecasting, and regression problems [2,3,4,5]. Among many algorithms used in ANN, the most popular training algorithms is back propagation (BP) algorithm.…”

Section: Introductionmentioning

confidence: 99%

An Improved Back Propagation Leaning Algorithm Using Second Order Methods with Gain Parameter

Nawi¹,

Saufi²,

Budiyono³

et al. 2018

IJIE

View full text Add to dashboard Cite

Back Propagation (BP) algorithm is one of the oldest learning techniques used by Artificial Neural Networks (ANN). It has successfully been implemented in various practical problems. However, the algorithm still faces some drawbacks such as getting easily stuck at local minima and needs longer time to converge on an acceptable solution. Recently, the introduction of Second Order Methods has shown a significant improvement on the learning in BP but it still has some drawbacks such as slow convergence and complexity. To overcome these limitations, this research proposed a modified approach for BP by introducing the Conjugate Gradient and Quasi-Newton which were Second Order methods together with 'gain' parameter. The performances of the proposed approach is evaluated in terms of lowest number of epochs, lowest CPU time and highest accuracy on five benchmark classification datasets such as Glass, Horse, 7Bit Parity, Indian Liver Patient and Lung Cancer. The results show that the proposed Second Order methods with 'gain' performed better than the BP algorithm.

show abstract

A Novel Method to Detect Segmentation points of Arabic Words using Peaks and Neural Network

Cited by 9 publications

References 16 publications

Online handwriting Arabic recognition system using k-nearest neighbors classifier and DCT features

Online handwriting Arabic recognition system using k-nearest neighbors classifier and DCT features

Redefining Selection of Features and Classification Algorithms for Room Occupancy Detection

An Improved Back Propagation Leaning Algorithm Using Second Order Methods with Gain Parameter

Contact Info

Product

Resources

About