Meskipun jumlah kelas salah, log berikut menunjukkan beberapa perbedaan kecepatan olah SVM antara GPU-GPU yang sudah saya coba. Hubungan yg nyata dg kecepatan adalah spek RAM, apakah DDR2 atau DDR3.
Using device 0: GeForce GTX 260
Input Train File Name: ../../data/shuttle/shuttle.scale
Input Test File Name: ../../data/shuttle/shuttle.scale.t
Code: All Vs All
Artificial Intelligence in Game: Emotional Respond from Indonesian Text using Text Classification and Fuzzy Logic
Abstract
Emotion plays a vital role in daily human communications. Many types of emotions such as joy, sadness, anger, fear have been known as important aspects of human behavior; however the role of emotion in human-computer interaction has not yet been much investigated.
This paper presents the role of emotion in NPC (Non Playable Characters) in game environment using text classification and fuzzy logic.
Keywords: game, artificial intelligence, text classification, fuzzy logic
Emotion Classification of Indonesian Text using Naive Bayes
TEKNOLOGI KREATIF
Membuat Film Animasi ala ITS
Jumat, 14 Agustus 2009 | 03:49 WIB
Film animasi kini semakin menjadi tren dunia. Selain tak perlu menyewa bintang film mahal, di film ini bisa berimajinasi liar, biaya produksinya juga relatif murah. Institut Teknologi Sepuluh Nopember, Surabaya, punya teknik tersendiri dalam membuat film animasi.
Read more…
The build process assumes that cuda is installed in /usr/local/cuda,
and that the CUDA SDK is installed at ~/cuda. If this is not the
case, set your CUDA_INSTALL_PATH and CUDA_SDK_PATH variables
appropriately.
another way round
==>
Read more…
install in the following order:
- driver (as root)
- toolkit (as root)
- sdk (as user)
Read more…
Internal Publication JSPS “Japan-East Asia Network of Exchange for Students and Youths-JENESYS Program” Report July 2009
Host University: Tohoku University, Japan
Host Professor: Tetsuo Kinoshita
URL: http://www.ka.riec.tohoku.ac.jp/index-e.html
Abstract
This paper briefly describes current development of a
life-like Indonesian conversational agent using machine learning
techniques. The key toward believable life-like agent is incorporating
emotions into an intelligent agent. In order to do that, our
text-based conversational agents need to classify text input given
by human user and respond emotionally correct.
However using available tools such as NB, MaxEnt and SVM,
our experiments for emotion-related text classification showed
accuracies below 60% which suggest the need of deeper investigations
on how to increase this accuracy. Regarding incorporating
emotions into an agent, some researchers have proposed to extend
rational agent’s BDI architecture. The experimental results are
promising and encourage us to explore further.
Index Terms—Life-like NPC, Conversational Agent, Emotion,
BDI
- http://search.cpan.org/~tpederse/Text-NSP-1.09/
- http://www.kwicfinder.com/kfNgram/kfNgramHelp.html
- http://homepages.inf.ed.ac.uk/lzhang10/ngram.html
Abstrak - Pengguna internet search engine semakin hari semakin banyak sejalan
dengan semakin beragamnya konten di internet. Dengan internet search engine,
pengguna internet akan mudah mencari konten internet sesuai dengan klasifikasi yang
diinginkan hanya dengan memasukkan kata kunci yang berkaitan dengan klasifikasi
konten tersebut. Didalam pemilihan kata kunci, pengguna awan lebih sering
memasukkan tujuan pencarian ke dalam kata kunci. Hal ini berbeda dengan pengguna
mahir yang memilih kombinasi kata kunci menggunakan reasoning berdasarkan
commonsense knowledge. Sebagai hasilnya, pengguna mahir lebih cepat menemukan
konten yang dibutuhkan.
Metode semantik diperkenalkan sebagai pemecahan atas masalah efektifitas
penggunaan internet search engine bagi pengguna awam. Dengan penggunaan
commonsense knowledge-base pada proses pencarian link website dalam klasifikasi
tertentu, diharapkan hasil pencarian antara pengguna awam mendekati kemampuan
pengguna mahir.
Kata Kunci – metode semantik, commonsense knowledge, internet search engine
Download Full paper in Indonesian, published at Seminar MMT ITS Jan 2009
Agen Percakapan Berbasis Pengetahuan Teks Berbahasa Indonesia
Abstract
Conversational agent is an intelligent agent utilizes natural language and computational linguistic techniques to engage user in human-like dialogue. In this research, we develop knowledge-based conversational agent which mined from free-text employs Information Retrieval (IR) modules. Example of the application of this research is conversational agent which has knowledge only in a certain domain, such as cashier agent used in Automated Teller Machine, librarian agent or tour guide agent in multimedia kiosk. Free text used is Indonesian narrative text. Experiments are conducted using geographical domain of tourism. Modules of IR can be deployed to enrich knowledge-base by human operator (user) with liberty of writing descriptions of tourism objects using Indonesian free-text. Knowlege-base uses network semantic which relates facts with semantic relations such as ”thing (isA, has)”, ”spatial(location)” and question words such as ”what(thing)”, ”where(location)”.
Keywords: conversational agent, information retrieval, knowledge-base, indonesian free-text.
fullpaper in Indonesian Agen Percakapan Berbasis Pengetahuan Teks Berbahasa Indonesia