Home | Bahasa | Sitemap | HelpDesk | UG-Pedia | Contact
Direktori UG : A - Z
Uzbek-Indonesian Joint International Conference, October 8-9, 2013  
Uzbek-Indonesian Joint International Conference On Economics And Management Toward Nation Character Development. Branch Of Russian Economic University After G.V.Plekhanov In Tashkent, Uzbekistan And ...

Indonesia Open Source Award 2012  
Indonesia Open Source Award (IOSA) 2012. For Further Information http://iosa.web.id/ ...

Home > Archive >
Thesis Defence of Doctoral in Information Technology: Miftah Andriansyah and Kemal Ade Sekarwati

On 7th May 2015, Doctoral Program Gunadarma University conducted doctoral thesis defence in Information Technology at auditorium of Gunadarma University, Campus D, Jl. Margonda Raya, Pondok Cina, Depok.

As the first candidat, Miftah Andriansyah presented his thesis dissertation with title "Annotated Corpus Based On Topic Models For Supporting Analysis in Client-Consuler Dialogue". The advisors committee are Profesor Doktor rer- nat Achmad Benny Mutiara Q. N., SKom, SSi as a Promotor and Doktor rer-nat. I Made Wiryana, SKom, SSi, MAppSc as a Co-Promotor, and as external examiner is Insinyur Paulus Insap Santosa M.Sc., Ph.D.

The main obstacle in this research on the specific domain of data analysis is the need for predefined corpus and the type of language. Problems frequently encountered by human coder in data analysis in psychotherapy is inconsistent and scalability. Topic models (TM) strategy is used as a decision support system in the search for topics or keywords that discover a theme or pattern of the content of the document.

TM-based Latent Dirichlet Allocation (LDA) is applied to calculate the probability of words that contribute in developing models of the topic. Tests performed on a single (snippet) document and the combined document of Consuler-Client Dialogue (CCD). The test results concluded: TM can predict the topic of the document; TM can be accepted by users as a predictive tool through an evaluation by an expert on the document snippet.

An annotated corpus is generated directly from the document in Indonesian language. This study also builds prototypes CCD analysis support system based on TM. Suggestions for future development to be considered in the testing, including: semantic factors, stemming process improving, parsing improvements, mistyping and the use of informal grammar (i.e slang)

The second candidate is Kemal Ade Sekarwati with Profesor Doktor Djati Kerami as The Promotor, Doktor Lintang Yuniar Banowosari and Doktor rer. nat. I Made Wiryana as CoPromotor, also Ir. Kridanto Surendro, MSc., PhD. as a external examiner. She defended her thesis dissertation with title "Combination Similarity Test for Different Type of Documents with Semantics"

Researches in Indonesian-language documents similarity test were previously done using Rabin-Karp algorithm, String Matching algorithm, Cosine Similarity calculation, and Dice Similarity. These experiments were using same type of documents, while similarity-test researches in different type of documents using semantics were limited. Researches in Indonesian-language documents similarity test using semantics in the field of education include automated essay grading and final reports.

Her research is focused on the similarity test of different types of documents using semantics. The results are shown in similarity percentages. The algorithm of documents-similarity test was done in five steps: (i) documents input, (ii) pre-processing, (iii) the calculation of term frequency-inverse document frequency (tf-idf), (iv) the calculation of latent semantic indexing (LSI), and (v) the calculation of documents similarity using cosine similarity, dice similarity, and Jaccard similarity.

The results using different types of documents by utilizing the topic modeling shows that the similarity of two different types of documents, successfully tested.

The  Senate Committee of Gunadarma University determine these two candidates was graduated with honors. Doctor Miftah Andriansyah and Kemal Ade Sekarwati are 73rd and 74th Doctor of Information Technology in Gunadarma University.

Expected from the results of these study can contribute and benefit theoretically and technology development.