Computer Science and Engineering

This paper deals with lexicon and system development for word segmentation in Bangla language.Our goal in this paper is to develop a morphological segmentation algorithm that can work well for Bangla and to address the problem of unsupervised word segmentation across different languages.From a hand-corrected Bangla corpus, 5000 popular words were segmented into suffixes, prefixes and roots manually.These were the sample lexicon used as seed for next step. A system was developed using C language to automate the Segmentation process based on hand made lexical database.The System was evaluated on several pages of Bangla text and achieved a success rate of about 83.05%.In our observation the system will work with full success if twice the volume of lexicon database and this system may have a huge impact particularly to learn and use Bangla for the people which will enhance their socio-economic life greatly.

Steganography program is mainly concerned about covering the way that a mystery message is being sent, just as hiding the substance of the message. Any computerized picture is contained pixels of various sizes of lattices; different picture steganography algorithms have been created. In this research, it has been exploited the irregular pixel selection based improved LSB exchange technique to hide a message into a cover image. We have considered a mystery message and a spread picture. In the LSB approach, the essential thought is to supplant the Least Significant Bits (LSB) of the spread picture with the bits of the message to be covered up without wrecking the property of the spread picture altogether. The LSB-based system is the most testing one as it is hard to separate between the spread item and stego-object as not many LSB bits of the spread article are supplanted. In digital image there are diverse types of file setup available such as Bitmap, PNG, JPEG, GIF etc. The proposed...

Since Bangladesh is an agrarian nation, its economy for the most part relies upon farming yield growth and social agro industry items. Agriculture is to a great extent affected by some profoundly flighty parameters such as temperature and rainwater in Bangladesh. Growth of agriculture also depends on weather parameters like temperature, rainfall, humidity as well as various soil parameters, soil moisture, surface temperature and crop rotation. Since, now-a day’s Bangladesh is quickly progressing into specialized advancement therefore technology will end up being beneficial to agriculture that will expand crop efficiency which brings about better productions to the farmers. One of those is also an essentially significant task in agricultures’ yield prediction. Before the cultivation process the research suggests different area based beneficial crops. It recommends some crops for a specific territory of land that are financially savvy for farming. Here, the study considered six main c...

In past few eras, emotion recognition from speech is one of the hottest research topic in the field of Human Computer Interaction. Many researches are going on various types of language, but for Bengali language, it is still very novice. In this work, 4 emotional state have been recognized i.e. happy, sad, angry and neutral from Bengali Speech Dataset. Proposed approach uses Pitch and Mel-frequency Cepstral Coefficient (MFCC) feature vectors to train k-Nearest Neighbor classifier for this work. A self-built Bengali emotional speech dataset has been used for both training and testing purpose. The dataset consists of consist of 50 people with 400 isolated emotional sentences. Using this dataset and above technique, we achieved 87.50% average accuracy rate, with detection accuracy each emotion (happy, sad, angry, neutral) respectively 80.00%, 75.00%, 85.00% and 75.00% in this work.

Disease samples are naturally fewer than healthy samples which introduces bias in the training of machine learning (ML) models. Current study focuses in learning discriminating patterns between cesarean and non-cesarean phenomena based on a dataset consisting of 161 features of total 692 cesarean and 5465 non-cesarean samples which comes as four folds based on four different hospitals (hospital A, B, C and D). The dataset is noisy, contains missing values, features are at different scales and above all, 161 features are quite a large in number and risks containing unnecessary information with respect to learning to separate the C-section class from non-cesarean.This study introduced a data pre-processing pipeline, resolving issues with data imbalance, handling missing values, identifying and deleting outliers, etc. A novel ensemble model is proposed which is able to consistently perform better irrespective of data volumes (data fold A, A+B, A+B+C and A+B+C+D) and pre-processing pipeline and achieved 96-99% accuracy across data volumes. Finally, the proposed model's decision-making was explained in terms of prominent features where higher values of features like Episiotomy, age of women and Fetal intrapartum pH accounts for causing C-section. INDEX TERMS Cesarean section, feature selection, data balancing, machine learning, explainable AI.

In this Fourth Industrial Revolution, the transport system is now at a different level. Electric Vehicle is now gaining importance day by day for their uncommon features. But, intrusion and unwanted attacks are one of the major problems of this kind of vehicle, and detecting the anomalies is an urgent task to run this smoothly. In this study, we proposed an ML-based intrusion detection system that finds the best ML algorithm for detection. We use Explainable AI tools to show how the features impact the performance of the classifiers. The proposed methodology includes missing value handling, outlier removal, normalization, and imbalanced data handling techniques in the preprocessing phase. We use Explainable AI (XAI) tools SHAP to show the feature importance and impact of the features. We use Logistic Regression, Decision Tree, Support Vector Classifier, Gradient Boosting Classifier, Random Forest (RF), Extreme Gradient Boosting, and K Nearest Neighbors Machine Learning (ML) algorithms for analysis purposes. Among all the algorithms, RF performs better than others; it shows 99.64% accuracy with 99% precision, recall, and F1 score. XAI tool SHAP is used on RF, and we get the model explainability.

Unlike English characters, one of the major drawbacks in recognizing handwritten Bengali script is the massive amount of characters in Bengali language and their complex shapes. There are 50 complex shaped characters in Bengali alphabet set and working with this huge amount of characters with an appropriate set of feature is a tough problem to solve. Moreover, the ambiguity and precision error are common in handwritten words. In addition, among the huge amount of complex shaped characters, some are very similar in shape those possess severe difficulty to recognize handwritten Bengali characters. Bearing in mind the complexity of the problem, an efficient approach for recognizing handwritten Bengali alphabet is proposed in this work. This proposed approach for identifying Bengali characters is based on character geometry-oriented feature extraction for different handwritten characters. In this paper, different image processing steps are used including image acquisition, digitization ...

Unlike English characters, one of the major drawbacks in recognizing handwritten Bengali script is the massive amount of characters in Bengali language and their complex shapes. There are 50 complex shaped characters in Bengali alphabet set and working with this huge amount of characters with an appropriate set of feature is a tough problem to solve. Moreover, the ambiguity and precision error are common in handwritten words. In addition, among the huge amount of complex shaped characters, some are very similar in shape those possess severe difficulty to recognize handwritten Bengali characters. Bearing in mind the complexity of the problem, an efficient approach for recognizing handwritten Bengali alphabet is proposed in this work. This proposed approach for identifying Bengali characters is based on character geometry-oriented feature extraction for different handwritten characters. In this paper, different image processing steps are used including image acquisition, digitization , preprocessing, segmentation and feature extraction for tackling the difficulty. Most importantly, the geometry based feature extraction method has been employed to extract the effective features from the Bengali characters for the classification purposes. Then, the classification result was measured for SVM and Artificial Neural Network (ANN) based classifiers on self-generated training and testing data sets which contain 2500 different samples of 50 characters in the Bengali character-set. The proposed technique produces an average recognition rate of 84.56% using SVM and 74.47% using ANN.

Unlike English characters, one of the major drawbacks in recognizing handwritten Bengali script is the massive amount of characters in Bengali language and their complex shapes. There are 50 complex shaped characters in Bengali alphabet set and working with this huge amount of characters with an appropriate set of feature is a tough problem to solve. Moreover, the ambiguity and precision error are common in handwritten words. In addition, among the huge amount of complex shaped characters, some are very similar in shape those possess severe difficulty to recognize handwritten Bengali characters. Bearing in mind the complexity of the problem, an efficient approach for recognizing handwritten Bengali alphabet is proposed in this work. This proposed approach for identifying Bengali characters is based on character geometry-oriented feature extraction for different handwritten characters. In this paper, different image processing steps are used including image acquisition, digitization ...