The rapid growth of the Internet has caused the increase in the amount of textual information available, such as in blogs, discussion forums and review sites on the web, where the texts surely have the emotion content. Emotion is one appearence of people behaviour and it is an important performance in human computer interaction (HCI). Human express the emotion in the form of facial expression, speech and writing text. Recently, researchers in computational linguistic (CL) areas are interested in the attention of emotion for Sentiment Analysis (SA). SA naturally observes the emotion conveyed by a text, and at the same time, distinguishing positive and negative valence. The wide areas of CL research, actually considerable for investigating the emotion dimension detection and searching the approaches and techniques in the term of emotion recognition (ER). There are two significant trends of research in the area, the emotion recognition based on state affective computing and the real time using brain signal machines. The two areas have the same aim for getting the improvement result in sentiment analysis with the mapping of emotion recognition provided. The exclusive work on emotion detection is comparatively rare and lacks empirical evaluation research. This paper provides the overview of past and recent research on emotion detection as well as some approaches and techniques used and shows the linked between both SA and ER.
This paper presents a work in developing a semantic-based question answering system (QAS) for Indonesian Translation of Quran (ITQ). This research is motivated by the lacks of previous built QAS that caused by a keyword-based retrieval. Instead of keeping the retrieval method, we shifted to a semantic approach where the retrieval process is done by using a semantic similarity measurement. In doing so, we built an ontology of ITQ to get the concepts as well as verses where they appear in. We applied three factoid question types on the QAS that including Who, Where, and When. Furthermore, a weighted vector for each concept that belongs to respective expected answering type (also called as named entity group) i.e. Person, Location, and Time is generated in order to feed semantic interpreter on user question. From 222 concepts defined from the ontology, we clustered them into 77, 24, and 6 concepts for Person, Location, and Time respectively. Since we found there are some characteristics of texts in ITQ, we developed our own modules to deal with including generate the inverted index and named entity recognition. Answer extraction is conducted by applying some features extraction in order to score the answer candidates. Evaluation of the system is designed by providing two data set of question and answer where the first one is purposed to measure the effectiveness of semantic approach comparing with keyword-based retrieval and the last one aims to know system performance in regard the appearance of concepts in ITQ.
15 September 2023 in Journal of Applied Data Sciences
This study aims to apply Convolutional Neural Networks (CNN) and image augmentation techniques in digit recognition using the MNIST dataset. We built a CNN model and experimented with various image augmentation techniques to improve digit recognition accuracy. The results showed that the use of CNN with image augmentation techniques was effective in improving digit recognition performance. In the data collection stage, we used the MNIST dataset consisting of images of handwritten digits as training and testing data. After building the CNN model, we apply image augmentation techniques such as rotation, shift, and flipping to the training data to enrich the data variety and prevent overfitting. The evaluation results show that the CNN model that has been trained with image augmentation techniques produces significant accuracy, with a maximum accuracy of 99.81%. We also performed an ensemble of several CNN models and found that this approach increased the digit recognition accuracy to 99.79%. This research has the potential for further development. Recommendations for further research include exploring more specific and complex image augmentation techniques, as well as using more challenging datasets. In addition, future research may consider improvements to the CNN architecture used or combining it with other methods such as recurrent neural networks (RNN).
This paper presents a work in generating Weighted Vector for each Concept in Indonesian Translation of Quran (ITQ). This task is done in aiming to provide a resource needed in implementing a semantic-based question answering system (QAS) for Indonesian ITQ, particularly in retrieving semantically related verses. Semantic approach on QAS employs Ontology concepts of the domain. Since there is no Ontology for ITQ remains, we built one by utilizing the existing Ontology from Quranic Arabic corpus (http://corpus.quran.com/). Furthermore, each leaf concept that enriched by related Quran verse (as its instance) had a representation vector of terms that occur in the corresponding Quran verse to express how strength the concept in relates with verse terms. This vector is assigned with a weight resulted from applying TFIDF method. From 222 leaf concepts in the Ontology, we applied the process only to those that categorized as a member group of Person, Location, and Time named entity. They are 107 in a total. The result shows that the most strength concept in association with verse terms is syaitan which is scored at 0.895 of 1. In overall, 16.82 % concepts had score that more than 0.4, following by 14.95%, 23.36% and 11.21% concepts scored at more than 0.3 ,0.2 and less than 0.1 respectively, and finally the rest ones were the biggest in volume where 33.64% concepts obtained score more than 0.1 and less than 0.2.
Zakat is the fourth pillar of Islam which is obligatory and must be implemented for those who can afford. Zakat maal is an obligation for those who have excess wealth of basic needs. Many Muslims are less so understand the charity and lack of socialization would be charity. This research aims to develop multi-platform zakat maal mobile web application using jQuery framework. It is built to deliver speed, stability, and an excellent cross-browser experience for web mobile visitor. Mobile Application Development Life Cycle (MADLC) includes the following phases: identification, design, development, prototyping, testing, deployment and maintenance, used to support developer create a mobile application rapidly.
Text Summerization, is a topic that is related to the fields of philosophy and linguistics are also included in the social sciences, so often sought by researchers in computer science who prefer something in the field of exact sciences. Interestingly, the Text Summerization applications, is a system that will make a summary or conclusion of tens or hundreds of texts with similar themes in an overview that produces new knowledge. This system becomes very important if you have a problem cases to take a series of conclusions from existing text. The purpose of this research is to develop knowledge-graph method to do case studies to determine whether the knowledge graph that can be used as an efficient instrument in analyzing a text with the same theme in large numbers, then the results become much simpler in the form of graphs that can be used as new knowledge.