Celebrity endorsement is a phenomenon in which companies advertises their products by using celebrity services, and celebrities take advantage of their popularity to promote a brand or product of the company through social media. In this study, KFC did a celebrity endorsement to make their menu more popular. KFC choose to work with Raditya Dika to promote their latest menu, KFC Salted Egg Chicken. This study will examine whether in such cases there is a change in public sentiment towards the product after the celebrity endorsement. It can be done using text mining and sentiment analysis. There are several algorithms that can be used to perform sentiment analysis, one of them is Support Vector Machine. Support Vector Machine (SVM) was chosen because this method is quite accurate in various studies. SVM also takes into account various features of the document, including features that often do not appear on the document, so it can reduce the loss of information from the data. The data used in this research are taken from YouTube and Twitter comment about KFC Salted Egg Chicken. Several step was done in this sentiment analysis research, that are preprocessing text, feature extraction, classification, and evaluation. The result model is tested and evaluated before and after endorsement by looking at the value of accuracy, precision, recall, and f1-measure. The test result of accuracy, precision, recall, and f-measure before endorsement were 67,83%, 69%, 68%, and 66%. After the endorsement, the test results were 74.06%, 74%, 74%, and 74% respectively. The results of this study indicate that SVM has an accurate measurement in sentiment analysis studies. Moreover, this study found that there was not significant change in public sentiment regarding the product before and after the celebrity endorsement.

An application can assist organizations in achieving the goals to be achieved by facilitating ongoing work processes. This happened in the Information Systems Study Program at one of the best private universities, namely Telkom University, where the SI Study Program has a website called PIPE and has one feature to be able to predict student graduation. However, this feature is currently being developed with an easy flow, so it requires development in the implementation of graduation achievements. Researchers solve these problems by building an assessment model based on academic data on the effect of choosing a specialization. Data mining is needed in this study to form predictive patterns, then one of the data mining groups is based on classification and using machine learning to perform automated assessments so that they can be sustainably performed. In determining the time and delay, using the decision tree method based on the C4.5 algorithm. The accuracy results obtained using the C4.5 algorithm are 94.11%, then the factor that becomes the root node is Jumlah SKS Lulus and the results have an influence on the selection of specialization. So that the results of this graduation model can be applied to the PIPE application. Keyword: C4.5 algorithm; classification; decision tree; graduation prediction Abstrak: Sebuah aplikasi dapat membantu organisasi dalam mencapai tujuan yang ingin dicapai dengan memudahkan proses kerja yang sedang berlangsung. Seperti yang terjadi pada Prodi Sistem Informasi yang ada pada salah satu Perguruan Tinggi Swasta terbaik yaitu Universitas Telkom, dimana pada Prodi SI memiliki website bernama PIPE dan memiliki salah satu fitur untuk dapat melakukan prediksi kelulusan mahasiswa. Namun fitur tersebut saat ini dikembangkan dengan alur penentuan sederhana, sehingga memerlukan pengembangan dalam hal implementasi algoritma prediksi kelulusan. Peneliti melakukan penyelesaian masalah tersebut dengan membangun model prediksi kelulusan berdasarkan rekam data akademik terhadap pengaruh pemilihan peminatan. Data mining dibutuhkan dalam penelitian ini untuk membentuk pola penyelesaian prediksi, kemudian salah satu pengelompokan data mining berdasarkan tugasnya adalah klasifikasi dan menggunakan machine learning untuk melakukan prediksi kelulusan secara otomatis terhadap data baru agar dapat dilakukan secara berkelanjutan. Dalam melakukan klasifikasi prediksi kelulusan tepat waktu dan terlambat, menggunakan metode decision tree berdasarkan algoritma C4.5. Hasil akurasi yang didapat dengan menggunakan algoritma C4.5 adalah sebesar 94,11%, kemudian faktor yang menjadi root node adalah Jumlah SKS Lulus dan hasil memiliki pengaruh terhadap pemilihan peminatan. Sehingga hasil model prediksi kelulusan ini dapat diterapkan pada aplikasi PIPE. Kata kunci: algoritma C4.5; decision tree; klasifikasi; prediksi kelulusan.

The learning process in online lectures through the Learning Management System (LMS) will produce a learning flow according to the event log. Assessment in a group of parallel classes is expected to produce the same assessment point of view based on the semester lesson plan. However, it does not rule out the implementation of each class to produce unequal fairness. Some of the factors considered to influence the assessment in the classroom include the flow of learning, different lecturers, class composition, time and type of assessment, and student attendance. The implementation of process mining in fairness assessment is used to determine the extent to which the learning flow plays a role in the assessment of ten parallel classes, including international classes. Moreover, a decision tree algorithm will also be applied to determine the root cause of the student assessment analysis based on the causal factors. As a result, there are three variables that have effects on student graduation and assessment, i.e attendance, class, and gender. The variable lecturer does not have much impact on the assessment but has an influence on the learning flow.

No abstract

Information Systems is one of the existing study program at Telkom University that has produced many graduates since it was established in 2008. However, not all graduates produced successfully completed the study period during the four years of normal study. The percentage of graduates on time has some decline between the target and the achievement of the study program. From academic year 2014/2015 to 2016/2017 decrease annually about 1% every year, which is it becomes problems for the credibility and existence of study program and also for academic planners who may have an impact on accreditation assessment process of the study program when it is audited. One of the efforts that can be done by the study program to increase the students on time graduation rate is by making decision support system dashboard that giving early warning to the lecturer or the head of the study program if there are students who are predicted not to graduate on time. By using the C4.5 algorithm to perform the data analysis by looking at the causes of student's graduation time and pureshare methodology to perform dashboard development method. The result of this study is a prototype of decision support system dashboard, because there are lack of analysis in decision making and the dashboard only showing information and temporary prediction. The data model that used on this research is labeling data that has been processed using C4.5 algorithm and data that has been through data cleansing process using Pentaho Data Integration. This prototype is expected to be used as a reference base to support academic planners in order to make this application run with real time data.