Data Mining - Concepts and Techniques.

Poetry, more than any other genre draws attentions of students and teachers. However, both students and teachers face challenges in reading and understanding poetry. Poetry for teaching English is widely adopted but understanding poetry is problematic. Reading strategies can address the problem. Therefore, this study investigated the strategies that students perceived they use, the actual strategies used and identified the different strategies that they used when they read Contemporary and Shakespearean poetries. In this research, the most common strategies that English as a Foreign Language (EFL) post-graduate students perceived they used, in comparison to what is actually used while they read poetry in English were studied. This study adopted the mixed method design which considers quantitative and qualitative data in the analysis. The instruments for data collection included a questionnaire (Survey of Poetry Reading Strategies) to identify the perceived strategies, and Think-aloud protocols to identify the actual strategies used while reading Contemporary and Shakespearean poetries. The participants were selected based on convenience random sampling. Sixty participants took part in the quantitative data collection and ten of them were purposely selected for qualitative data collection. Results from the questionnaire showed use of Problem solving strategies such as re-reading was perceived to be used more often than Global and Support strategies such as analysing and evaluating. In addition, results from the protocol showed that Global strategies were used more than the Support and Problem solving strategies. Analysis of the two types of poetry used in the study showed that readers did not use different strategies in reading them. As a conclusion, the results from the qualitative data showed that Global strategies are generally used for reading poetry. It is recommended that future research examines these strategies for reading of poetry from different era.

Strategies used when reading contemporary and shakespearean poetries by english as a foreign language students

Proteins are biological polymers of amino acid residues. Pro- teins perform various functions within living organisms. Multiple pro- teins carry out these tasks by forming functional modules. Each func- tional module possesses community structure. For identifying functional modules, a lot of community detection or clustering algorithms were de- signed, but most of those algorithms suer by inappropriate clustering results which do not make any sense biologically. Though some of the al- gorithms came out with better results but too high time complexity was not of great help. Recently an ecient algorithm was designed which out- performed other existing algorithms, named Fast Protein Network Clus- tering or FPNC algorithm. We have worked on that algorithm and im- proved its performance by introducing the concept of second neighbours (neighbours of neighbours of any vertex), named as Second order Fast Protein Network Clustering algorithm or 2nd order FPNC algorithm. By coming up with the concept of 2nd neighbours, 2nd order FPNC algo- rithm has better scoring function and better functional module mapping results indicating ecient identication

/pdf/introducing-the-concept-of-second-neighbours-to-fpnc-4wtf72bykf.pdf

Introducing the Concept of Second Neighbours to FPNC algorithm for Improving the Functional Modules Detection

Poetry for teaching English is widely adopted but understanding poetry is problematic. Reading strategies can address this problem. Therefore, in this research, the most common strategies that Iranian post-graduate students perceive that they use were studied. This study adopted the quantitative method design in data analysis. The instrument for data collection is a questionnaire ( Survey of Poetry Reading Strategies or SPRS). The participants were selected based on convenience random sampling. Sixty participants took part in the quantitative data collection. Results from the questionnaire showed use of Problem Solving strategies such as re-reading was perceived to be used more often than Global strategies such as making judgment and opinion and Support strategies such as analysing and evaluating . This study contributes the useful reading strategies in reading poetry by Iranian students. These results can be useful for both students and teachers in reading poetry. Article visualizations:

Iranian students' perceptions on poetry reading strategies

Exploring Awareness of Online Reading Strategies Used by EFL Learners in a Developing Country: A Study on Undergraduate Students in Bangladesh.

First of all, I would like to thank to almighty Allah for giving me the strength to work on
this thesis. At the same time, I would like to express my gratitude to CSE Department of
MIST for giving me an opportunity to work for this thesis as a part of the degree.
Special and heartfelt thanks to my supervisor, Dr. Md. Abul Kashem Mia for encouraging
me to work with this interesting subject. Despite of busy schedule, his regular supervision,
guidance and dedicated supporting role is worth to mention. In preparing the report, he
actively supported me and reviewed the report painstakingly. I thankfully acknowledge his
contribution to enrich the report.
I would like to acknowledge, with utmost gratitude, the significant contribution of Mr.
Swapnil Saha, in data analysis part. Heartfelt thanks to Dr. Gazi Nurun Nahar Sultana,
Chief Scientist, Genetic Engineering and Biotechnology Research, Centre for Advanced
Research in Sciences (CARS), University of Dhaka for giving domain expert opinions
regarding the implications of such research findings. Special thanks to Dr. Md. Shamsur
Rahman, Monash University, for his sincere support, suggestion and continuous
encouragement in completing the work. I also wish to express my sincerest appreciation to
Prof. Dr. Md. Shamsul Arefin, CUET, Assoc Prof Nazrul and Lt Cdr Anis, MIST,
Senior Lec Md. Sarwar Kamal, EWU, Doctor Ratan, BSM Medical University and Prof Dr
Altaf Hussain , BSMRMU. Their sincere and wholehearted cooperation, consultation and
opinions along with supporting information helped me a lot in doing the thesis work.
Thanks to my service staffs specially Nimai Chandro for giving clerical support.
And at last, my heartfelt thanks to my wife who gave me constant support and
encouragement in the successful completion of this work.

/pdf/frequent-contiguous-pattern-mining-over-biological-sequences-4mzb00kyzz.pdf

Frequent contiguous pattern mining over biological sequences of protein misfolded diseases.

Proteins are responsible for health disorder, evolution of species. These tasks are carried out by forming functional modules. Each functional module possesses community structure. For identifying the functional modules, a lot of community detection or clustering algorithms are designed, but most of the algorithms suffer by either high computational time or inappropriate clustering results. We propose an algorithm based on relative vertex-to-vertex clustering value and agglomerative hierarchical method, known as Fast Protein Network Clustering algorithm. It is faster than existing algorithms and resolves the most common clustering problem-clustering any vertex (protein) of degree one from its neighbour. We also tested our algorithm with the most popular four algorithms in respect to functional module mapping and efficiency analysis. FPNC algorithm successfully outperformed existing algorithms which is the current state-of-the-art agglomerative approach to functional module identification.

Fast protein network clustering algorithm: a new approach for clustering protein-protein interaction networks to detect functional modules

There is no doubt that the biggest challenge Bangladesh has to face in this millennium is to find successful solutions of the problem of poverty, by improving the human development conditions, especially in rural areas. It is believed that Information and Communication Technologies (ICTs) can play a vital role in achieving this objective. ICT is gaining much attention in present times for its ability towards human development and poverty alleviation. In this paper, the prospect of rural human development opportunities in Bangladesh has been explored from the viewpoint of Community Informatics (CI) model.

https://papers.ssrn.com/sol3/Delivery.cfm/SSRN_ID1619770_code1334827.pdf?abstractid=1619770&mirid=1

Mohammad Shahedul Islam

Papers

Exploring Awareness of Online Reading Strategies Used by EFL Learners in a Developing Country: A Study on Undergraduate Students in Bangladesh.

Frequent contiguous pattern mining over biological sequences of protein misfolded diseases.

Fast protein network clustering algorithm: a new approach for clustering protein-protein interaction networks to detect functional modules

The Potentiality of ICTs in Rural Human Development in Bangladesh and a Conceptual Framework of Community Informatics to Facilitate the Development