Layer-based Composite Reputation Bootstrapping

We propose a novel generic reputation bootstrapping framework for composite services. Multiple reputation-related indicators are considered in a layer-based framework to implicitly reflect the reputation of the component services. The importance of an indicator on the future performance of a component service is learned using a modified Random Forest algorithm. We propose a topology-aware Forest Deep Neural Network (fDNN) to find the correlations between the reputation of a composite service and reputation indicators of component services. The trained fDNN model predicts the reputation of a new composite service with the confidence value. Experimental results with real-world dataset prove the efficiency of the proposed approach.

Examination of the Correlation Between 8th grade Students’ Cyber Bully/Victim Behaviors and their Self-confidence Values

In the study, the correlation between cyberbullying and cyber victimization, which is included in the digital literacy skill in the Social Studies Curriculum, and self-confidence value was examined. From this point of view, the research aims to determine the degree and direction of the correlation between 8th grade students cyberbully/victim behaviors and their self-confidence value. As a data collection tool; The CyberBully/Victim Scale developed by Ayas and Horzum (2010); The Self-Confidence Scale developed by Akin (2007), and the Personal Information Form developed by the researcher were used to access personal information. The research sample consists of a total of 455 students studying at the 8th grade of public and private schools in the Nilufer, Yildirim, and Kestel districts of Bursa. Spearman’s Rho Correlation analysis was used in the analysis of the data. As a result of the study, a statistically significant moderate-level inverse correlation was found between students cyberbully/victim status and their tendency levels of self-confidence.

Thermography signs of lung inflammation

The purpose of the study was to find out if infrared thermography of the thorax is the method to select the patients with lung inflammationMaterial, methods: Thermograms were accumulated and processed in the «TVision» cloud storage («Dignosis», Russia). Special regions of interest (ROI) were automatically created: 1. on the front and back of the thorax roughly in the projection of the upper lobe (ULP) and the lower lobe (LLP) of the lung; 2.e lines on the front surface of the thorax. Two types of temperature gradients were calculated: between ULP and LLP (by subtraction mean temperature in LLP from mean temperature in ULP) (ΔT1); between both ULP and both LLP on the back of the thorax (ΔT2). Approximation confidence value for the polynomial trend line (R²) along the marked lines on the front surface of the thorax also calculated. Totally 489 thermograms, were analyzed, included 337 from healthy patients (group 1) and 152 from patients with confirmed diagnosis of lung inflammation (group 2)Results: R² value was higher in the group 1 compare to group 2 (0.58 ± 0.16 vs 0.3 ± 0.2, p < 0.05). ΔT1 value was negative only in patients from group 2, as well as ΔT1 value greater than 0.4 °C.Conclusion: three independent thermographic criteria suitable for detecting lung inflammation were found, so infrared thermography is the valuable method for screening this pathology.

Penentuan Paket Promosi Pakaian PT. D&C Production dengan Menggunakan Algoritma FP-Growth

PT. D&C Production is a center for selling, foam mattresses, mattresses of all sizes and models, may types of car accessories and various kinds of women's and men's underwear. Problems regarding the decline in sales resulted in the accumulation of goods so that it became a loss. Data mining can be a solution to overcome these problems. This study will use the fp-growth algorithm to form an association model with the aim of helping companies increase their sales by creating underwear promotional packages. The data set that will be used to support this research is the sales transaction data set for the period April 2020 to December 2020. The results show that known rules have been obtained using the fp-growth algorithm, where the rules of this association can create strategies to increase clothing sales. In the form of five association rules that are ready to be used for making clothing promotional packages by meeting the support values and confidence values that have been set at the beginning, namely having a confidence value above 80% and a support value above 25%.

Application of Association Rule Method Using Apriori Algorithm to Find Sales Patterns Case Study of Indomaret Tanjung Anom

Data mining can generally be defined as a technique for finding patterns (extraction) or interesting information in large amounts of data that have meaning for decision support. One of the well-known and commonly used association rule discovery data mining methods is the Apriori algorithm. The Association Rule and the Apriori Algorithm are two very prominent algorithms for finding a number of frequently occurring sets of items from transaction data stored in databases. The calculation is done to determine the minimum value of support and minimum confidence that will produce the association rule. The association rule is used to produce the percentage of purchasing activity for an itemset within a certain period of time using the RapidMiner software. The results of the test using the priori algorithm method show that the association rule, that customers often buy toothpaste and detergents that have met the minimum confidence value. By searching for patterns using this a priori algorithm, it is hoped that the resulting information can improve further sales strategies.

Implementasi Algoritma FG-Growth untuk Sistem Rekomendasi Penjualan Produk

The variety of stationery marketed, makes business competition increasingly fierce in order to provide the best service to customers. Abundant sales transaction data, triggering piles of data so that it requires data mining processing techniques, namely association rule mining using the FP-Growth algorithm. Algorithm that generates frequent itemset used in the process of determining the rules that can produce an option by taking a product sales transaction data object. The test results show a rule that has the best confidence value and lift ratio of 100%, as well as 80% support with the rules that every purchase of a ballpoint product can be sure to buy a notebook from the dataset used as a sample data in the system trial (50 names). goods and 7 transaction data) with minimum support (5% = 0.05) and minimum confidence (30% = 0.3).

A CNN-Based Wearable Assistive System for Visually Impaired People Walking Outdoors

In this study, we propose an assistive system for helping visually impaired people walk outdoors. This assistive system contains an embedded system—Jetson AGX Xavier (manufacture by Nvidia in Santa Clara, CA, USA) and a binocular depth camera—ZED 2 (manufacture by Stereolabs in San Francisco, CA, USA). Based on the CNN neural network FAST-SCNN and the depth map obtained by the ZED 2, the image of the environment in front of the visually impaired user is split into seven equal divisions. A walkability confidence value for each division is computed, and a voice prompt is played to guide the user toward the most appropriate direction such that the visually impaired user can navigate a safe path on the sidewalk, avoid any obstacles, or walk on the crosswalk safely. Furthermore, the obstacle in front of the user is identified by the network YOLOv5s proposed by Jocher, G. et al. Finally, we provided the proposed assistive system to a visually impaired person and experimented around an MRT station in Taiwan. The visually impaired person indicated that the proposed system indeed helped him feel safer when walking outdoors. The experiment also verified that the system could effectively guide the visually impaired person walking safely on the sidewalk and crosswalks.

Data to Knowledge-Based Transformation

The objectives of this study is to analyze a large amount of data that often appears to create a knowledge base that can be utilized by firm to enhance their decision support system. The authors used the association rules with rapid miner software, data mining approach, and predictive analysis that contains various data exploration scenarios. The study provides important evidence for adopting data mining methods in the industrial sector and their advantages and disadvantages. Chevron Pacific Indonesia (CPI) has a type of computer maintenance activity. Currently, a numerous errors often occur due to the accuracy in computer maintenance which has a major impact on production results. Therefore, this study focuses on association rules using growth patterns that often appear on variables that have been determined into the algorithm (FP-growth) which results in knowledge with a 100% confidence value and a 97% support value. The value results of this study has support and trust are expected to become knowledge for top management in deciding evergreen IT-business routines.

IMPLEMENTASI DATA MINING MENGGUNAKAN METODE ALGORITMA APRIORI PADA TOKO DHURROH ELEKTRONIK

In the business world, every store must of course be able to compete and think about how the store can continue to grow and be able to expand its business scale. In order to increase sales of products sold, business actors must have various strategies. One way is by utilizing all sales transaction data that has occurred in the store itself. Dhurroh Elektronik store is a store that sells various kinds of goods such as cellphone accessories. Management of sales data in this store is still done manually, namely by recording sales data in the sales book or sometimes when serving purchases just remembering it. The obstacle faced is that it is difficult to find out where the goods are not in accordance with the behavior of consumers' habits in buying goods at the same time. Based on the above problems, it is necessary to have a calculation to group data items based on their tendencies that appear together in a transaction with Data Mining calculations using the Apriori Algorithm method. The results of the calculation of the items that are most in-demand are if you buy a headset, you will buy a lamp with a 100% confidence value and 19% support, if you buy a radio you will buy a lamp with a 71% confidence value and 16% support, if you buy a data cable, you will buy a flashlight. with a 71% Confidence value and 16% Support, If you buy a battery, you will buy a Flashlight with a 71% Confidence value and 16% Support. Keywords: Data Mining, Apriori Algorithm..

THE OPTIMIZE OF ASSOCIATION RULE METHOD FOR THE BEST BOOK PLACEMENT PATTERNS IN LIBRARY: A MONTHLY TRIAL

Data mining is the process of finding interesting patterns and knowledge from large amounts of data. Sources of information service, especially in the library, include books, reference books, serials, scientific gray literature (newsletters, reports, proceedings, dissertations, theses, and others). The importance of this research being carried out in the library in this study aims to implement data mining with the association rule method to solve problems, especially in the placement of shelves based on the category of the printed version of the book collection. This research method uses a qualitative research approach. Data was collected using documentation techniques and deep analysis of existing weaknesses to identify user needs whose information was obtained through observation and interviews with key informants (admin, user, etc.). For example, the determination of the best book placement patterns can be done by looking at the results of the tendency of visitors to borrow books based on a combination of 2 item sets with 60 percent of confidence value every month or week and must be evaluated or take a calculate again.