فهرست مطالب

Journal of Computer and Knowledge Engineering
Volume:4 Issue: 1, Winter-Spring 2021

  • تاریخ انتشار: 1401/05/23
  • تعداد عناوین: 3
|
  • Sareh Mostafavi, Bahareh Pahlevanzadeh *, MohammadReza Falahati Qadimi Fumani Pages 1-10

    Automatic text classification, which is defined as the process of automatically classifying texts into predefined categories, has many applications in our everyday life and it has recently gained much attention due to the in-creased number of text documents available in electronic form. Classifying News articles is one of the applications of text classification. Automatic classification is a subset of machine learning techniques in which a classifier is built by learning from some pre-classified documents. Naïve Bayes and k-Nearest Neighbor are among the most common algorithms of machine learning for text classification. In this paper, we suggest a way to improve the performance of a text classifier using Mutual information and Chi-square feature selection algorithms. We have observed that MI feature selection method can improve the accuracy of Naïve Bayes classifier up to 10%. Experimental results show that the proposed model achieves an average accuracy of 80% and an average F1-measure of 80%.

    Keywords: Automatic Persian text classification, k-Nearest Neighbor, Naïve Bayes, News text classification, Text mining
  • Mahbobe Dadkhah, Saeed Araban *, Samad Paydar Pages 11-23

    Software testing is one of the most important activities for ensuring quality of software products. It is a complex and knowledge-intensive activity which can be improved by reusing tester knowledge. Generally, testing web applications involve writing manual test scripts, which is a tedious and labor-intensive process. Manually written test scripts are valuable assets encapsulating the knowledge of the testers. Reusing these scripts to automatically generate new test scripts can improve the effectiveness of software testing and reduce the cost of required manual interventions. In this paper, a semantic web enabled approach is proposed for automatically adapting and generating test scripts; it reduces the cost of human intervention across multiple scripts by accumulating the human knowledge as semantic annotations on test scripts. This is supported by designing an ontology which defines the concepts and relationships required for test script annotation. The proposed approach is based on novel algorithms for adapting and generating new test scripts. The initial experiments show that the proposed approach is promising as it successfully increases the level of test automation.

    Keywords: Automated testing, semantic web, Test adaptation, Test generation, Test ontology
  • Masoud Mahjoubi, Morteza Dadashi, Kooroush Manochehri *, Saadat Pourmozafari Pages 25-34
    Full adder is one of the essential circuits among the various processing elements used in VLSI and other technologies circuits, because they are mainly employed in other arithmetic circuits, such as multi-digit adders, subtractors, and multipliers. This paper proposes two efficient ternary full adders based on Carbon Nanotube Field-Effect Transistor (CNFET) technology. Using the adjustable nanotube diameter in CNFETs, these adders utilize arbitrary threshold voltages so that arithmetic operations can be performed with a radix of 3. For performance analysis, the proposed adder circuits are simulated in HSPICE with 32nm CNFET technology. In these simulations, different inputs are applied at different frequencies with different load capacitances placed at the output. Simulation results have shown that the proposed adders not only improve the speed, power consumption, and Power Delay Product (PDP) of the existing state-of-the-art designs but also improve the design complexity by reducing the number of transistors contained within the circuit.
    Keywords: CNFET, Ternary Adder, Multi-Value Logic, Ternary Logic, nanotechnology