The ease of sending data with the development of internet technology technology is now a concern, especially the problem of data confidentiality, integrity and information security. Cryptography is one of the techniques used to maintain data confidentiality and information security, the application of cryptographic techniques for information security and data integrity is highly dependent on the formation of keys. In this study proposed a frequency analysis approach to measure the level of information security of blowfish encryption results to determine the distribution form of each character used in the text and find out the exact frequency of each character used in the test text data. The encryption algorithm and description of blowfish method against plaintext are proven to be accurate, but the longer the key character used will greatly affect the level of information security that came  from encryption process, this is based on the results of the frequency analysis conducted.

Delta compression uses the previous block of bytes to be used as a reference in the compression process for the next blocks. This approach is increasingly ineffective due to the duplication of byte sequences in modern files. Another delta compression model uses the numerical difference approach of the sequence of bytes contained in a file. Storing the difference value will require fewer representation bits than the original value. Base + Delta is a compression model that uses delta which is obtained from the numerical differences in blocks of a fixed size. Developed with the aim of compressing memory blocks, this model uses fixed-sized blocks and does not have a special mechanism when applied to file compression in general. This study proposes a compression model by developing the concept of Base+Delta encoding which aims to be applicable to all file types. Modification and development carried out by adopting a dynamic block size using a sliding window and block header optimization on compressed and uncompressed blocks giving promising test results where almost all file formats tested can be compressed with a ratio that is not too large but consistent for all file formats where the ratio compression for all file formats obtained between 0.04 to 12.3. The developed compression model also produces compression failures in files with high uncompressed blocks where the overhead of additional uncompressed blocks of information causes files to become larger with a negative ratio obtained of -0.39 to -0.48 which is still relatively small and acceptable.

ASCII differentiation is a compression method that utilizes the difference value or the difference between the bytes contained in the input character. Technically, the ASCII differentiation method can be done using a coding dictionary or using windowing block instead of the coding dictionary. Previous research that has been carried out shows that the ASCII differentiation compression ratio is good enough but still needs to be analyzed on performance from the perspective of the compression ratio of the method compared to other methods that have been widely used today. In this study an analysis of the comparison of the ASCII Difference method with other compression methods such as LZW will be carried out. The selection of LZW itself is done by reason of the number of data compression applications that use the method so that it can be the right benchmark. Comparison of the compression ratio performed shows the results of ASCII differentiation have advantages compared to LZW, especially in small input characters. Whereas in large input characters, LZW can optimize the probability of pairs of characters that appear compared to ASCII differentiation which is glued to the difference values ​​in each block of input characters so that in large size characters LZW has a greater compression ratio compared to ASCII differentiation.

Pendekatan Data Science (ilmu data) memberi peluang besar untuk menggunakan data history dan mengubahnya menjadi wawasan yang berguna untuk membangun model prediksi penjualan masa depan. akan tetapi, model prediksi membutuhkan analisis data tertentu untuk menghasilkan model yang kuat, termasuk jumlah pelanggan, jumlah pelanggan yang hilang, tingkat penjualan rata-rata serta kecenderungan musiman. Makalah ini menganalisis data penjualan menggunakan kerangka kerja ilmu data dengan desain sesuai prinsip CRIS-DM yang terdiri dari tahapan pemahaman bisnis, pemahaman data, persiapan data, pemodelan, evaluasi, dan penerapan. Pemodelan digunakan algoritma Machine Learning untuk memprediksi penjualan di masa depan yang hasil kinerjanya dievaluasi dengan RMSE, MEA dan R^2. Berdasarkan hasil pengujian algoritma XGBoost dan LightGBM menghasilkan nilai R^2 mencapai 60% dengan tingkat kesalahan deteksi terendah dibandingkan algoritma lainnya..