A Food Recommender Based on Frequent Sets of Food Mining Using Image Recognition

Food recommendation is an important service in our life. To set a system, we searched a set of food images from social network which were shared or reviewed on the web, including the information that people actually chose in daily life. In the field of representation learning, we proposed a scalable architecture for integrating different deep neural networks (DNNs) with a reliability score of DNN. This allowed the integrated DNN to select a suitable recognition result obtained from the different DNNs that were independently constructed. The frequent set of foods extracted from food images was applied to Apriori data mining algorithm for the food recommendation process. In this study, we evaluated the feasibility of our proposed method.

Integrated Survival Analysis and Frequent Pattern Mining for Course Failure-Based Prediction of Student Dropout

A data-driven method to identify frequent sets of course failures that students should avoid in order to minimize the likelihood of their dropping out from their university training is proposed. The overall probability distribution of the dropout is determined by survival analysis. This result can only describe the mean dropout rate of the undergraduates. However, due to the failure of different courses, the chances of dropout can be highly varied, so the traditional survival model should be extended with event analysis. The study paths of students are represented as events in relation to the lack of completing the required subjects for every semester. Frequent patterns of backlogs are discovered by the mining of frequent sets of these events. The prediction of dropout is personalised by classifying the success of the transitions between the semesters. Based on the explored frequent item sets and classifiers, association rules are formed providing the estimates of the success of the continuation of the studies in the form of confidence metrics. The results can be used to identify critical study paths and courses. Furthermore, based on the patterns of individual uncompleted subjects, it is suitable to predict the chance of continuation in every semester. The analysis of the critical study paths can be used to design personalised actions minimizing the risk of dropout, or to redesign the curriculum aiming the reduction in the dropout rate. The applicability of the method is demonstrated based on the analysis of the progress of chemical engineering students at the University of Pannonia in Hungary. The method is suitable for the examination of more general problems assuming the occurrence of a set of events whose combinations may trigger a set of critical events.

MOOC opinion mining based on attention alignment

Purpose The purpose of this paper is to propose an attention alignment method for opinion mining of massive open online course (MOOC) comments. Opinion mining is essential for MOOC applications. In this study, the authors analyze some of bidirectional encoder representations from transformers (BERT’s) attention heads and explore how to use these attention heads to extract opinions from MOOC comments. Design/methodology/approach The approach proposed is based on an attention alignment mechanism with the following three stages: first, extracting original opinions from MOOC comments with dependency parsing. Second, constructing frequent sets and using the frequent sets to prune the opinions. Third, pruning the opinions and discovering new opinions with the attention alignment mechanism. Findings The experiments on the MOOC comments data sets suggest that the opinion mining approach based on an attention alignment mechanism can obtain a better F1 score. Moreover, the attention alignment mechanism can discover some of the opinions filtered incorrectly by the frequent sets, which means the attention alignment mechanism can overcome the shortcomings of dependency analysis and frequent sets. Originality/value To take full advantage of pretrained language models, the authors propose an attention alignment method for opinion mining and combine this method with dependency analysis and frequent sets to improve the effectiveness. Furthermore, the authors conduct extensive experiments on different combinations of methods. The results show that the attention alignment method can effectively overcome the shortcomings of dependency analysis and frequent sets.

Credit Card User Frequent Buying Predicton Analysis using Cluster Methods

Today the world becomes more digital. The cashless transactions are increased in all sectors. The large amounts of data in digital form are generated every day. The companies need to analyze the existing transactions, to predict the user requirements in the future. The payment during the purchase can be done in different modes by the user. In this work, the credit card transactions are analyzed. There are many data mining techniques are used to predict the frequent sets of items during purchase. Data clustering in one of the familiar and widely used technique to identify a similar set of items in a group or dataset. In this work, the two familiar existing techniques k-means and k-mediods are compared with the same datasets. The results show the best clustering algorithm.

Shorter but More Frequent Rest Periods: No Effect on Velocity and Power Compared to Traditional Sets not Performed to Failure

Performing traditional sets to failure is fatiguing, but redistributing total rest time to create short frequent sets lessens the fatigue. Since performing traditional sets to failure is not always warranted, we compared the effects of not-to-failure traditional sets and rest redistribution during free-weight back squats in twenty-six strength-trained men (28 ± 5.44 y; 84.6 ± 10.5 kg, 1RM-to-body-mass ratio of 1.82 ± 0.33). They performed three sets of ten repetitions with 4 min inter-set rest (TS) and five sets of six repetitions with 2 min inter-set rest (RR6) at 70% of one repetition maximum. Mean velocity (p > 0.05; d = 0.10 (-0.35, 0.56)) and mean power (p > 0.05; d = 0.19 (-0.27, 0.64)) were not different between protocols, but the rating of perceived exertion (RPE) was less during RR6 (p < 0.05; d = 0.93 (0.44, 1.40)). Also, mean velocity and power output decreased (RR6: 14.10% and 10.95%; TS: 17.10% and 15.85%, respectively) from the first repetition to the last, but the percentage decrease was similar (velocity: p > 0.05; d = 0.16 (0.30, 0.62); power: p > 0.05; d = 0.22 (-0.24, 0.68)). These data suggest that traditional sets and rest redistribution maintain velocity and power output to a similar degree when traditional sets are not performed to failure. However, rest redistribution might be advantageous as RR6 displayed a lower RPE.

DEVELOPING FISCAL DATA PROCESSING ANALYTICAL SYSTEM

The paper deals with the task of designing an analytical system for processing fiscal data. From a business point of view, such a system should solve the problem of analyzing a market basket, that is, finding the most typical patterns of purchases. From the point of view of data mining, the task of searching for association rules is solved, which consists of two stages: the search for all frequent sets with their support values and the acquisition of association rules based on the sets found. The first stage is provided by various search algorithms for frequent sets. In the paper, the algorithm chosen is the Frequent Pattern Growth Strategy (FPG) as the optimal one. The mathematical formulation of the task and the method for implementing the selected algorithm within the target system are given. The result of the work is a description of the fault‑tolerant and scalable model of the analytical system.

Indexing Arabic texts using association rule data mining

Purpose The purpose of this paper is to propose a new model to enhance auto-indexing Arabic texts. The model denotes extracting new relevant words by relating those chosen by previous classical methods to new words using data mining rules. Design/methodology/approach The proposed model uses an association rule algorithm for extracting frequent sets containing related items – to extract relationships between words in the texts to be indexed with words from texts that belong to the same category. The associations of words extracted are illustrated as sets of words that appear frequently together. Findings The proposed methodology shows significant enhancement in terms of accuracy, efficiency and reliability when compared to previous works. Research limitations/implications The stemming algorithm can be further enhanced. In the Arabic language, we have many grammatical rules. The more we integrate rules to the stemming algorithm, the better the stemming will be. Other enhancements can be done to the stop-list. This is by adding more words to it that should not be taken into consideration in the indexing mechanism. Also, numbers should be added to the list as well as using the thesaurus system because it links different phrases or words with the same meaning to each other, which improves the indexing mechanism. The authors also invite researchers to add more pre-requisite texts to have better results. Originality/value In this paper, the authors present a full text-based auto-indexing method for Arabic text documents. The auto-indexing method extracts new relevant words by using data mining rules, which has not been investigated before. The method uses an association rule mining algorithm for extracting frequent sets containing related items to extract relationships between words in the texts to be indexed with words from texts that belong to the same category. The benefits of the method are demonstrated using empirical work involving several Arabic texts.

Effects of Cluster Sets and Rest-Redistribution on Mechanical Responses to Back Squats in Trained Men

Eight resistance-trained men completed three protocols separated by 48-96 hours. Each protocol included 36 repetitions with the same rest duration, but the frequency and length of rest periods differed. The cluster sets of four (CS4) protocol included 30 s of rest after the 4th, 8th, 16th, 20th, 28th, and 32nd repetition in addition to 120 s of rest after the 12th and 24th repetition. For the other two protocols, the total 420 s rest time of CS4 was redistributed to include nine sets of four repetitions (RR4) with 52.5 s of rest after every four repetitions, or 36 sets of single repetitions (RR1) with 12 s of rest after every repetition. Mean (MF) and peak (PF) force, velocity (MV and PV), and power output (MP and PP) were measured during 36 repetitions and were collapsed into 12 repetitions for analysis. Repeated measures ANOVA 3 (protocol) x 12 (repetition) showed a protocol x repetition interaction for PF, MV, PV, MP, and PP (p-values from <0.001 to 0.012). No interaction or main effect was present for MF. During RR1, MV, PV, MP, and PP were maintained, but decreased throughout every 4-repetition sequence during CS4 and RR4. During CS4 and RR4, PF was less following a rest period compared to subsequent repetitions, whereas PF was maintained during RR1. These data indicate that rest redistribution results in similar average kinetics and kinematics, but if total rest time is redistributed to create shorter but more frequent sets, kinetics and kinematics may remain more constant.

Interactive Visual Analytics of Big Data

High volumes of a wide variety of data can be easily generated at a high velocity in many real-life applications. Implicitly embedded in these big data is previously unknown and potentially useful knowledge such as frequently occurring sets of items, merchandise, or events. Different algorithms have been proposed for either retrieving information about the data or mining the data to find frequent sets, which are usually presented in a lengthy textual list. As “a picture is worth a thousand words”, the use of visual representations can enhance user understanding of the inherent relationships among the mined frequent sets. However, many of the existing visualizers were not designed to visualize these mined frequent sets. This book chapter presents an interactive next-generation visual analytic system. The system enables the management, visualization, and advanced analysis of the original big data and the frequent sets mined from the data.