No abstract

Aims: The use of social media in generation Y and generation Z as a forum for communication is commonly used in everyday life. The use of language as a means of communication in social media also attracts attention, especially the use of Indonesian, as a national language and as a unifying language of the nation, Indonesian must of course be maintained. Especially in writing and using it. The frequency of writing words on social media often becomes a problem in Indonesian as a national language. Both consciously and unconsciously, Indonesian writing is often found that is not in accordance with the structure of grammar and vocabulary in social media, causing concerns about the use of Indonesian in the future that has strayed far from the general guidelines of Indonesian spelling. Researchers conduct research related to the use of basic words in social media. Study Design: The analysis process is carried out using an approach that starts from data collection, stemming and grouping data from random social media (WhatsApp, Instagram and Facebook). Place and Duration of Study: Data from the top 3 (three) social media (WhatsApp, Instagram and Facebook) with the keyword "belajar" between December 2019 and January 2020. Methodology: From the data collected there were 40492 basic words that were successfully obtained. Then the data obtained is done by sorting and deleting repeated words, so as to get the base word data of 5243 words. Results: From the results of the study obtained information that the word "ajar" is the most frequent word with 2502 times repeated. The words "yang" and the word "ga" are non-standard words most often appear with the appearance 759 and 530 times. Mistakes most often occur because of abbreviated writing and hypercorrection. Conclusion: The use of words that are not standard and not in accordance with the general guidelines of Indonesian spelling are words that are dominated by abbreviated words. Abbreviations like "yg" should "yang" rank first, while the word "ga" who should "tidak" rank second. The writing of words caused by hypercorrection is also one of the causes of errors in writing words. Hypercorrection words that appear include "zenius", "coba" and "smoga".

No abstract

2 ema.u@amikom.ac.id, 3 wa2n@akprind.ac.id AbstrakKeamanan dalam desaian suatu database merupakan hal yang sangat penting untuk diterapkan dalam membangun sebuah sistem informasi. Sering kali dalam implementasi keamanan database tidak begitu diperhatikan, baik dari segi tipe data, panjang data, maupun paramter yang berkaitan dengan transaksi data. Kesalahan dalam desain database biasanya baru akan disadari pada saat aplikasi sudah selesai dan sudah digunakan. Untuk memastikan sebuah sistem informasi berjalan dengan lancar maka akan digunakan analisa data berdasarkan rentang data tertinggi dan data terendah. Dengan menggunakan analisa rentang data maka data yang nilainya lebih rendah dari paremeter yang ditentukan akan dirubah ke nilai terendah yang sesuai dengan paramter. Begitu juga data yang nilainya lebih tinggi dari paramter yang telah ditentukan akan dirubah ke nilai tertinggi yang sesuai dengan parameter. Dari hasil pengujian yang dilakukan sebelum adanya filter check dan patameter tipe data dengan memasukan data yang berupa huruf, angka minus serta angka yang melebihi batas atas pada kolom nilai, data masih bisa tersimpan ke database. Pengujian selanjutnya dilakukan dengan memasukan angka -2,-3,-4,-5,-7,-7 serta angka diatas ambang batas atas 12,13,14,15,16,17, dari hasil pengujian yang dilakukan dapat disimpulkan bawah sebuah database dengan fungsi trigger jauh lebih aman dibandingkan dengan database yang hanya menerapkan fungsi pembatasan berdasarkan tipe datanya saja. AbstractSecurity in the design of a database is very important to be applied in building an information system. Often in the implementation of a database security is not given much attention, both in terms of data types, length, and parameters relating to data transactions. Errors in database design are usually only noticed when the application is complete and has been used. To ensure an information system runs well, data analysis will be used based on the highest and lowest data ranges. By using data range analysis, data whose value is lower than the specified parameter will be changed to the lowest value in accordance with the parameters. Data whose value is higher than predetermined parameters will be changed to the highest value in accordance with parameters. Results of tests conducted before the filter check and data type patameter by entering data in the form of letters, minus numbers and numbers that exceed the upper limit in the value column, data can still be saved to database. Further testing is done by entering the numbers -2, -3, -4, -5, -7, -7 and numbers above the upper threshold of 12,13,14,15,16,17, from some of the results of tests conducted can be concluded below a database with a trigger function is more secure than a database that only applies restrictions based on its data type.

Mango is a popular fruit for local consumption and export commodity. Currently, Indonesian mango export at 37.8 M accounted for 0.115% of world consumption. Pest and disease are the common enemies of mango that degrade the quality of mango yield. Specialized treatment in export destinations such as gamma-ray in Australia, or hot water treatment in Korea, demands pest-free and high-quality products. Artificial intelligence helps to improve mango pest and disease control. This paper compares the deep learning model on mango fruit pests and disease recognition. This research compares Visual Geometry Group 16 (VGG16), residual neural network 50 (ResNet50), InceptionResNet-V2, Inception-V3, and DenseNet architectures to identify pests and diseases on mango fruit. We implement transfer learning, adopt all pre-trained weight parameters from all those architectures, and replace the final layer to adjust the output. All the architectures are re-train and validated using our dataset. The tropical mango dataset is collected and labeled by a subject matter expert. The VGG16 model achieves the top validation and testing accuracy at 89% and 90%, respectively. VGG16 is the shallowest model, with 16 layers; therefore, the model was the smallest size. The testing time is superior to the rest of the experiment at 2 seconds for 130 testing images.