TEXT MINING DALAM MEMBANDINGKAN METODE NAÏVE BAYES DENGAN C.45 DALAM MENGIDENTIFIKASI BERITA HOAX PADA MEDIA SOSIAL

Hoax news (hocus to trick) has a very big influence in disseminating information, especially in the world of social media. News has an important impact on social and political conditions, and news can move the economy of a country. For this reason, it is necessary to have an analysis to classify hoax news and not hoaxes, and have high accuracy in classifying the news. In this study, two methods were used as a comparison in achieving high accuracy, namely the Naïve Bayes method which is famous for having high accuracy in classification with little data, and the C.45 method which can minimize noise in the data. The data used are 300 articles with 10 topics which contain hoax and non-hoax news. The data is obtained from the internet through social media, such as Twitter, Instagram and Facebook. Testing using the Naïve Bayes method has a higher accuracy than the C.45 method. The amount of data used has a major influence on the test results, if more data enters the training stage, then this study will have higher accuracy. However, the results of this test can be recommended to increase accuracy in the construction of a hoax news detection system.

Application of Nave Bayes Algorithm for SMS Spam Classification Using Orange

Mobile phones have become a necessity for everyone. SMS is a communication service that is used to send and receive short messages in the form of text on mobile phones. Among all the advantages of SMS, there is a very annoying activity called spam (unsolicited commercial advertisements). Spam is the continuous use of electronic devices to send messages. called spammers. Spam messages are sent by advertisers with the lowest operating costs. Therefore, there are a lot of spammers and the number of messages requested is huge. Therefore, many aspects are harmed and disturbed. When SMS enters the user's mobile device, this study aims to classify spam and ham SMS. SMS classification adopts naive Bayes method. By looking at the contents of the SMS, the application of the naive Bayes method in data mining can distinguish unwanted SMS from non-spam. Results The classification accuracy rate is 0.999%. Based on the research that I have done, the Naive Bayes method can classify 1000 SMS spam data contained in the SMS spam data set file correctly.

Comparison of Classification Data Mining C4.5 and Naïve Bayes Algorithms of EDM Dataset

The purpose of this research is to choose the best method by comparing two classification methods of data mining C4.5 and Naïve Bayes on Educational Data Mining, in which the data used is student graduation data consisting of 79 records. Both methods are tested for validation with 10-ford X Validation and perform a T-Test difference test to produce a table that contains the best method ranking. Different results were obtained for each method. Based on the results of these two methods, it is very influential on the dataset and the value of the area under curve in the Naïve Bayes method is better than the C4.5 method in various datasets. Comparison of the method with the 10-Ford X Validation test and the T-Test difference test is that the Naïve Bayes method is better than C4.5 with an average accuracy value of 73.41% and an under-curve area of 0.664.

Aksi Penyerangan Non-Player Character (NPC) Menggunakan Metode Naive Bayes pada Shooter Game

<p>Aksi penyerangan pada <em>non-player character </em>(NPC) merupakan salah satu substansi penting dalam pembuatan <em>game</em>. Dalam melakukan penyerangan diperlukan strategi khusus agar NPC tidak mudah dikalahkan. Salah satunya adalah adanya variasi serangan terhadap pemain. Salah satu metode yang digunakan untuk mengatur penyerangan NPC adalah <em>rulebase</em>. Metode <em>rulebase </em>dapat memberikan variasi serangan sesuai kondisi NPC, namun metode <em>rulebase </em>bisanya menghasilkan <em>behaviour </em>yang statis dan tidak adaptif jika terdapat kondisi baru. AI seperti ini akan mudah diprediksi dan repetitif sehingga menurunkan tingkat tantangan bermain <em>game</em>. Untuk mengatasi masalah tersebut banyak peneliti yang menggunakan teknik <em>learning</em>. Salah satunya menggunakan metode <em>naïve bayes.</em> Pada penelitian ini dilakukan penerapan metode <em>naïve bayes </em>sebagai strategi penyerangan NPC pada <em>shooter game</em>. Metode <em>naïve bayes </em>digunakan untuk keputusan serangan yang diambil oleh NPC. Adapun parameter yang digunakan untuk keputusan penyerangan adalah nyawa, jarak, jumlah granat, dan jumlah amunisi yang dimiliki NPC. Sedangkan keputusan penyerangan dibagi menjadi serangan tembak, serangan granat, dan serangan pisau. Hasil penelitian menunjukkan penerapan metode <em>naïve bayes </em>membuat NPC mampu melakukan penyerangan secara otonom jika terdapat kondisi baru dengan akurasi 80%. Penerapan metode <em>naïve bayes </em>juga lebih unggul dalam persentase kemenangan NPC dibanding metode <em>rulebase</em>. Tingkat kemenangan NPC menggunakan metode <em>naïve bayes </em>sebesar 60% sedangkan <em>rulebase </em>sebesar 16%.</p><p> </p><p><em><strong>Abstract</strong></em></p><p><em>Non-Player Character’s (NPC) attacks behaviour is one important substance in making games. While NPC attacks needs specific strategy to not get defeated easily. One of the NPC attacks strategy is a variation of offense to player. One of the methods to manage the NPC attack strategy is rulebase. Rulebase method can give variations of the NPC attacks according in conditions, but rulebase method usually producing static behaviour and not adaptive where there is new condition. AI like this would easy too predictive and repetitive so that decrease the challenge of playing games. To overcome these problems, we use naïve bayes method. In this study, naïve bayes method are applied as an NPC’s attack strategy to the shooter game. Naïve bayes method used for attack decisions taken by the NPC. The parameters used for the attack’s decision are health point, distance, number of grenades, and number of ammunitions owned by the NPC. While attacks decision is divided into firing attacks, grenade attacks, and melee attacks. The results showed that the use naïve bayes method can attack autonomously if there are new condition with an accuracy of 80%. The implementation of naïve bayes method at NPC more superior than rulebase method in percentage of NPC winning. The NPC win rate uses the naïve bayes method is 60% while the rulebase is 16%.</em><em></em></p><p><em><strong><br /></strong></em></p>

KLASIFIKASI PEMBAGIAN ARUS LALU LINTAS MENGGUNAKAN ALOGARITMA NAÏVE BAYES DAN MODEL LINEAR

The density of traffic flow is a problem for every big city, especially as it is easy to have a private vehicle, causing the flow to increase every year. So to overcome traffic flow, a system that can make optimal traffic performance is needed is needed. The purpose of this study is to determine whether the road conditions are empty, smooth, dense and very congested so as to produce a prediction of road options whether to continue passing the road or find another way, as well as to test the accuracy of traffic flow using the naive bayes method and the liner model. The classification stages carried out are data input, data preprocessing, classification, and the results of accuracy, precision, and recall. And the results of this study the naive bayes method obtained higher accuracy than the linear model, namely for naive bayes accuracy 95.70%, precision 95.67%, and recall 100%, while for naive bayes accuracy 92.10%, precision 95.68%, and recall 96.20%. then the result is the naive bayes method is superior in the traffic flow data classification process. And the results of decision making obtained results from traffic flow data obtained that the road is empty so that the road can be passed without having to find another way. Keywords - Classification, Naive Bayes, Traffic, Linear Model, Flow Density AbstrakKepadatan arus lalu lintas menjadi masalah setiap kota-kota besar, apalagi seiring mudah nya dalam memiliki kendaraan pribadi sehingga menimbulkan arus yang meningkat pada setiap tahunnya. Maka untuk penanggulangan arus lalu lintas dibutuhkan sistem yang bisa membuat kinerja lalu lintas yang optimal. Tujuan penelitian ini adalah mengetahui kondisi jalan apakah lengang, lancar, padat dan sangat padat sehingga menghasilkan prediksi opsi jalan apakah tetap melewati jalan tersebut atau mencari jalan lain, serta menguji tingkat akurasi arus lalu lintas menggunakan metode naive bayes dan model liner. Dengan tahapan klasifikasi yang dilakukan yaitu input data, preprocessing data, klasifikasi, dan hasil accuracy, precision, dan recall. Dan hasil penelitian ini metode naive bayes mendapatkan accuracy lebih tinggi dari model linier yaitu untuk naive bayes accuracy 95.70%, precision 95.67%, dan recall 100%, sedangkan untuk naive bayes accuracy 92.10%, precision 95.68%, dan recall 96.20%. maka hasilnya metode naive bayes lebih unggul dalam proses klasifikasi data arus lalu lintas. Dan hasil dari pengambilan keputusan didapat hasil dari data arus lalu lintas didapatkan jalan tersebut lengang sehingga jalan tersebut dapat dilalui tanpa harus mencari jalan lain. Kata Kunci - Klasifikasi, Naive Bayes, Lalu Lintas, Model Linier, Kepadatan Arus