{"title":"KPK操作分析总结将手工操作引入算法支持向量机、朴素贝叶斯、BERBASIS粒子群优化","authors":"Hernawati Hernawati, Windu Gata Kedua","doi":"10.30998/faktorexacta.v12i3.4992","DOIUrl":null,"url":null,"abstract":"It is known from various public sentiments conveyed through comments on social media twitter against the capture operations carried out by the corruption eradication commission (KPK) that currently it does not meet the expectations of the community, where officials who are only officials have small corruption rates, not corruption As for the classification algorithms that have strong accuracy at this time are Support Vector Machine and Naïve Bayes algorithms, calculation of Support Vector Machine method for tweet data from 78 positive tweet data and 78 negative tweet data, resulting in an accuracy of 80.77% and AUC 0.867. Whereas the results of accuracy with the Naïve Bayes method are 76.92% and AUC 0.729. Having a difference in accuracy of 3.3%, and after optimizing with the Operator Vector Machine (PSO) weight Particle Swarm Optimization the accuracy is 83.79% and AUC 0.910, while for Naïve Bayes (PSO) produces an accuracy of 80.13% and AUC 0.771 Has a difference in accuracy of 3.6%.Diketahui dari berbagai sentimen masyarakat yang disampaikan melalui komentar di media sosial twiter terhadap operasi tangkap tangan yang dilakukan oleh Komisi Pemberantasan Korupsi (KPK) nyatanya saat ini belum memenuhi harapan masyarakat, dimana pejabat yang di ott hanya pejabat yang mempunyai angka korupsi kecil, bukan korupsi yang besar adapun algoritma klasifikasi yang kuat akurasinya saat ini adalah algoritma Support Vector Machine untuk data tweet dari 78 data tweet positif dan 78 data tweet negatif, menghasilkan akurasi sebesar 80.77% dan AUC 0.867. Sedangkan hasil akurasi dengan metode Naïve Bayes adalah 76.92% dan AUC 0.729. Memiliki selisih akurasi sebesar 3.3%, dan setelah di optimalisasi dengan oprator Weight Partical Swarm Optimization untuk Support Vector Machine (PSO) menghasilkan akurasi 83.79% dan AUC 0.910, sedangkan untuk Naïve Bayes (PSO) menghasilkan akurasi sebesar 80.13% dan AUC 0.771 memiliki selisih akurasi sebesar 3.6%.","PeriodicalId":53004,"journal":{"name":"Faktor Exacta","volume":" ","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2019-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"11","resultStr":"{\"title\":\"SENTIMEN ANALISIS OPERASI TANGKAP TANGAN KPK MENURUT MASYARAKAT MENGGUNAKAN ALGORITMA SUPPORT VECHTOR MACHINE, NAÏVE BAYES, BERBASIS PARTICLE SWARM OPTIMIZITION\",\"authors\":\"Hernawati Hernawati, Windu Gata Kedua\",\"doi\":\"10.30998/faktorexacta.v12i3.4992\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"It is known from various public sentiments conveyed through comments on social media twitter against the capture operations carried out by the corruption eradication commission (KPK) that currently it does not meet the expectations of the community, where officials who are only officials have small corruption rates, not corruption As for the classification algorithms that have strong accuracy at this time are Support Vector Machine and Naïve Bayes algorithms, calculation of Support Vector Machine method for tweet data from 78 positive tweet data and 78 negative tweet data, resulting in an accuracy of 80.77% and AUC 0.867. Whereas the results of accuracy with the Naïve Bayes method are 76.92% and AUC 0.729. Having a difference in accuracy of 3.3%, and after optimizing with the Operator Vector Machine (PSO) weight Particle Swarm Optimization the accuracy is 83.79% and AUC 0.910, while for Naïve Bayes (PSO) produces an accuracy of 80.13% and AUC 0.771 Has a difference in accuracy of 3.6%.Diketahui dari berbagai sentimen masyarakat yang disampaikan melalui komentar di media sosial twiter terhadap operasi tangkap tangan yang dilakukan oleh Komisi Pemberantasan Korupsi (KPK) nyatanya saat ini belum memenuhi harapan masyarakat, dimana pejabat yang di ott hanya pejabat yang mempunyai angka korupsi kecil, bukan korupsi yang besar adapun algoritma klasifikasi yang kuat akurasinya saat ini adalah algoritma Support Vector Machine untuk data tweet dari 78 data tweet positif dan 78 data tweet negatif, menghasilkan akurasi sebesar 80.77% dan AUC 0.867. Sedangkan hasil akurasi dengan metode Naïve Bayes adalah 76.92% dan AUC 0.729. Memiliki selisih akurasi sebesar 3.3%, dan setelah di optimalisasi dengan oprator Weight Partical Swarm Optimization untuk Support Vector Machine (PSO) menghasilkan akurasi 83.79% dan AUC 0.910, sedangkan untuk Naïve Bayes (PSO) menghasilkan akurasi sebesar 80.13% dan AUC 0.771 memiliki selisih akurasi sebesar 3.6%.\",\"PeriodicalId\":53004,\"journal\":{\"name\":\"Faktor Exacta\",\"volume\":\" \",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2019-11-22\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"11\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Faktor Exacta\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.30998/faktorexacta.v12i3.4992\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Faktor Exacta","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.30998/faktorexacta.v12i3.4992","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 11
摘要
从社交媒体推特上对肃贪委抓捕行动的各种评论中可以得知,目前肃贪委的抓捕行动并没有达到社会的期望,只做官员的官员腐败率小,不腐败。目前准确率较强的分类算法是支持向量机和Naïve贝叶斯算法。用支持向量机方法对78条正推文数据和78条负推文数据进行推文数据的计算,得到准确率为80.77%,AUC为0.867。而Naïve贝叶斯方法的准确率为76.92%,AUC为0.729。准确率相差3.3%,使用算子向量机(PSO)加权粒子群优化后的准确率为83.79%,AUC为0.910,而对于Naïve,贝叶斯(PSO)产生的准确率为80.13%,AUC为0.771,准确率相差3.6%。这句话的意思是:“我的意思是说,我的意思是说,我的意思是说,我的意思是说,我的意思是说,我的意思是说,我的意思是说,我的意思是说,我的意思是说,我的意思是说,我的意思是说,我的意思是说,我的意思是说,我的意思是说,我的意思是说,我的意思是说,我的意思是说,我的意思是说,我的意思是说,我的意思是说,支持向量机(Support Vector Machine)将数据推文的78个数据推文为正,78个数据推文为负,得到的数据推文的80.77%和AUC为0.867。Sedangkan hasil akurasi dengan方法Naïve Bayes adalah 76.92%, AUC 0.729。权重粒子群优化算法支持向量机(PSO) menghasilkan akurasi 83.79%和AUC 0.910, Bayes (PSO) menghasilkan akurasi 80.13%和AUC 0.771 Memiliki selisih akurasi sebesar 3.6%。
SENTIMEN ANALISIS OPERASI TANGKAP TANGAN KPK MENURUT MASYARAKAT MENGGUNAKAN ALGORITMA SUPPORT VECHTOR MACHINE, NAÏVE BAYES, BERBASIS PARTICLE SWARM OPTIMIZITION
It is known from various public sentiments conveyed through comments on social media twitter against the capture operations carried out by the corruption eradication commission (KPK) that currently it does not meet the expectations of the community, where officials who are only officials have small corruption rates, not corruption As for the classification algorithms that have strong accuracy at this time are Support Vector Machine and Naïve Bayes algorithms, calculation of Support Vector Machine method for tweet data from 78 positive tweet data and 78 negative tweet data, resulting in an accuracy of 80.77% and AUC 0.867. Whereas the results of accuracy with the Naïve Bayes method are 76.92% and AUC 0.729. Having a difference in accuracy of 3.3%, and after optimizing with the Operator Vector Machine (PSO) weight Particle Swarm Optimization the accuracy is 83.79% and AUC 0.910, while for Naïve Bayes (PSO) produces an accuracy of 80.13% and AUC 0.771 Has a difference in accuracy of 3.6%.Diketahui dari berbagai sentimen masyarakat yang disampaikan melalui komentar di media sosial twiter terhadap operasi tangkap tangan yang dilakukan oleh Komisi Pemberantasan Korupsi (KPK) nyatanya saat ini belum memenuhi harapan masyarakat, dimana pejabat yang di ott hanya pejabat yang mempunyai angka korupsi kecil, bukan korupsi yang besar adapun algoritma klasifikasi yang kuat akurasinya saat ini adalah algoritma Support Vector Machine untuk data tweet dari 78 data tweet positif dan 78 data tweet negatif, menghasilkan akurasi sebesar 80.77% dan AUC 0.867. Sedangkan hasil akurasi dengan metode Naïve Bayes adalah 76.92% dan AUC 0.729. Memiliki selisih akurasi sebesar 3.3%, dan setelah di optimalisasi dengan oprator Weight Partical Swarm Optimization untuk Support Vector Machine (PSO) menghasilkan akurasi 83.79% dan AUC 0.910, sedangkan untuk Naïve Bayes (PSO) menghasilkan akurasi sebesar 80.13% dan AUC 0.771 memiliki selisih akurasi sebesar 3.6%.