Classification Of Attributes Research Articles

Attribute Classification and Reduct Computation in Multi-Adjoint Concept Lattices

The problem of reducing information in databases is an important topic in formal concept analysis, which has been studied in several articles. In this article, we consider the fuzzy environment of the multi-adjoint concept lattices, since it is a general fuzzy framework that allows us to easily establish degrees of preference on the elements of the considered database. We introduce algorithms to discover the information contained in the relational system. By means of these algorithms, we classify the attributes of a multi-adjoint context, and build a minimal subset of attributes preserving the information of the original knowledge system.

A Multi-Task Framework for Facial Attributes Classification through End-to-End Face Parsing and Deep Convolutional Neural Networks.

Human face image analysis is an active research area within computer vision. In this paper we propose a framework for face image analysis, addressing three challenging problems of race, age, and gender recognition through face parsing. We manually labeled face images for training an end-to-end face parsing model through Deep Convolutional Neural Networks. The deep learning-based segmentation model parses a face image into seven dense classes. We use the probabilistic classification method and created probability maps for each face class. The probability maps are used as feature descriptors. We trained another Convolutional Neural Network model by extracting features from probability maps of the corresponding class for each demographic task (race, age, and gender). We perform extensive experiments on state-of-the-art datasets and obtained much better results as compared to previous results.

Machine Learning Based Approach on Food Recognition and Nutrition Estimation

Abstract Nowadays, standard intake of healthy food is necessary for keeping a balanced diet to avoid obesity in the human body. In this paper, we present a novel system based on machine learning that automatically performs accurate classification of food images and estimates food attributes. This paper proposes a deep learning model consisting of a convolutional neural network that classifies food into specific categories in the training part of the prototype system. The main purpose of the proposed method is to improve the accuracy of the pre-training model. The paper designs a prototype system based on the client server model. The client sends an image detection request and processes it on the server side. The prototype system is designed with three main software components, including a pre-trained CNN model training module for classification purposes, a text data training module for attribute estimation models, and a server-side module. We experimented with a variety of food categories, each containing thousands of images, and through machine learning training to achieve higher classification accuracy.

АНАЛІЗ ПОНЯТІЙНО-ТЕРМІНОЛОГІЧНИХ СИСТЕМ СУСПІЛЬНОЇ ГЕОГРАФІЇ ТА ВЗАЄМОЗВ’ЯЗКІВ МІЖ НИМИ

The purpose of this research work is to identify and analyse the most important conceptual and terminological systems of social geography and the relationships between them. Method. General scientific methods, including analysis, scientific synthesis, analytical method, methods of comparison and generalization were used in the study. Accordingly, the need to identify the category of monodevelopment and its inherent differences in the theory of social geography requires a detailed consideration of its etymology, semantic separation from other categories, i.e. scientific justification considering the specifics of geographical understanding. In the course of the research the method of classification had been used in the separation of the constituent elements of the territories of monodevelopment. The scientific novelty lies in the disclosure of the conceptual and terminological apparatus of social geography and its structuring into a number of conceptual and terminological systems, within which the content of concepts is deepened, the terms denoting certain concepts are specified, in particular the concept of «monodevelopment» in socio-geographical science, taking into account, first of all, the attributes of the territory (properties, relations, processes). It has been revealed that monodevelopment is aimed at the formation of territorial social systems of the highest hierarchical level, in the structure of which individual elements will begin their own monodevelopment, which will end with differentiation or integration of subsystems of the territorial social system. The main attributes of the territory in the interpretation of monodevelopment in social geography had been analyzed. Based on the classification of attributes of the territory (properties, relations, processes) formed in the theory of social geography, the author singled out new types of monodevelopment of the territory with increasing weight of one element, such as monoresource, monodemographic, monofunctional, monocentric development and others. Practical meaning. The results of this study contribute to a deeper socio-geographical understanding of the processes of monodevelopment, their genesis, features of their course and provide an opportunity to model the long-term development of territorial social systems, to achieve expected results due to long-term transformation. They can be used for further study of monosystems of different hierarchical levels, as well as for the development of practical recommendations and programs for the development of individual monoterritories. It is especially important to clarify the category of «monoterritory» and the separation of monoterritories as special territorial formations, which will further determine aspects of the development of this type of territorial formation. It is extremely important in today's decentralization reform in Ukraine, when new united territorial communities are formed; many of which are economically monodirectional.

Dermoscopic attributes classification using deep learning and multi-task learning

Abstract Melanoma is a serious condition affecting people around the world. Therefore, automatic diagnostics of pigmented skin lesions is an area of high interest. Possible approach to diagnostics is based on detection of dermoscopic attributes. In this study we consider developing an end-to-end solution for the task of dermoscopic attributes detection based on the data of ISIC (International Skin Imaging Collaboration) competition dataset. We investigate the possibility to improve performance using labeling of different modality in the form of segmentation masks. We propose to use the segmentation task as an auxiliary task and allow sharing knowledge between these two tasks by training a neural network of Y-Net architecture, which allows sharing of the weights between segmentation and classification branches. Additionally, we propose several modes of combining training of Y-Net based on segmentation and classification labels. The best of proposed methods has reached F1 score of 0.527 comparing to plain classification’s 0.502.

PARADIGM OF SPONGE CITY CONSTRUCTION BASED ON REGIONAL CLASSIFICATION

Based on the systematic analysis of natural factors such as hydrogeology and climate conditions of Chinese cities, and the social factors such as new and old urban areas and land use function types, this paper classifies and summarizes the pilot experience of sponge cities in China, and puts forward the construction strategy and paradigm of sponge cities based on regional classification in China and social attributes classification in cities, so as to provide references for sponge city construction promotion in the future.

Multi Labeled Imbalanced Data Classification Based on Advanced Min-Max Machine Learning

Some true applications, for example, content arrangement and sub-cell confinement of protein successions, include multi-mark grouping with imbalanced information. Different types of traditional approaches are introduced to describe the relation of hubristic and undertaking formations, classification of different attributes with imbalanced for different uncertain data sets. Here this addresses the issues by utilizing the min-max particular system. The min-max measured system can break down a multi-mark issue into a progression of little two-class sub-issues, which would then be able to be consolidated by two straightforward standards. Additionally present a few decay procedures to improve the presentation of min-max particular systems. Trial results on sub-cellular restriction demonstrate that our strategy has preferable speculation execution over customary SVMs in settling the multi-name and imbalanced information issues. In addition, it is additionally a lot quicker than customary SVMs

Intelligent Bimodal Segmentation Applied on CFA Images of Food Products

Images segmentation is an essential step in the non-destructive analysis of food products by image analysis. It participates upstream in the operations of classification, identification or extraction of image attributes. To implement our segmentation approach, we extracted from each CFA image
 The R, G, B (for the RGB color space), and H, S, V (for the HSV color space). We then calculate the histogram of each of the color components. The component S for which the bimodal histogram was the most pronounced was selected and the binary mask was created from this component. The created binary mask has been filtered using morphological closing and opening operations. The application of each binary mask to its image CFA host, allowed us to separate with success the pixels representing the food products from the pixels of the background of the image CFA.

What's in the box?! Towards explainable machine learning applied to non-residential building smart meter classification

Feature engineering and data-driven classification models are at the forefront of analysis of large temporal sensor data from the built environment. In previous efforts, temporal features were engineered from the whole building hourly electrical meter data from 507 non-residential buildings. These features fall within the three general categories of statistics, model, and pattern-based and can be used to identify various behavior in the structure of the whole building electrical meter data. In this paper, a deeper investigation is made of exactly what types of behavior are most important in the context of two classification scenarios: the primary use of a building and the level of performance the building has when compared to its peers. The highly comparative time-series analysis (hctsa) toolkit is used to analyze the most important temporal features for the classification of various building performance attributes. In the first analysis, a comparison is made to distinguish the behavior between university dormitories (70 buildings) and laboratories (95 buildings) as an example of interpreting the classification of the primary-use-type of a building. In the second analysis, a comparison of buildings with high (165 buildings) versus low (169 buildings) consumption is used to extract and understand the behavior that indicates the level of the energy performance of a building. These two case study examples provide a foundation for further explainable machine learning techniques in both classification and prediction as applied to buildings. This effort is the first example of machine learning with an explicit focus on the interpretability of classification for smart meter data from non-residential buildings.

Drinks & Crowds

The design of computational methods to recognize alcohol intake is a relevant problem in ubiquitous computing. While mobile crowdsensing and social media analytics are two current approaches to characterize alcohol consumption in everyday life, the question of how they can be integrated, to examine their relative value as informative of the drinking phenomenon and to exploit their complementarity towards the classification of drinking-related attributes, remains as an open issue. In this paper, we present a comparative study based on five years of Instagram data about alcohol consumption and a 200+ person crowdsensing campaign collected in the same country (Switzerland). Our contributions are two-fold. First, we conduct data analyses that uncover temporal, spatial, and social contextual patterns of alcohol consumption on weekend nights as represented by both crowdsensing and social media. This comparative analysis provides a contextual snapshot of the alcohol drinking practices of urban youth dwellers. Second, we use a machine learning framework to classify individual drinking events according to alcohol and non-alcohol categories, using images features and contextual cues from individual and joint data sources. Our best performing models give an accuracy of 82.3% on alcohol category classification (against a baseline of 48.5%) and 90% on alcohol/non-alcohol classification (against a baseline of 65.9%) using a fusion of image features and contextual cues in this task. Our work uncovers important patterns in drinking behaviour across these two datasets and the results of study are promising towards developing systems that use machine learning for self-monitoring of alcohol consumption.

Lithofacies Control on Reservoir Quality of the Viola Limestone in Southwest Kansas and Unsupervised Machine Learning Approach of Seismic Attributes Facies-Classification

The hydrocarbon development of the Viola Limestone in southern Kansas, USA, has encountered challenges, regarding the development of a robust data-based model of the reservoir-quality controls. The legacy understanding that hydrocarbon entrapment and reservoir-quality are controlled by structure, has resulted in less than optimal drilling results. In this study, an integration of petrographic and geophysical well-logs analyses established the main reservoir quality control as dolomitization-induced porosity. The dolomitization control is supported by comparing best-fit trends on density-porosity well log values with typical model-trends of limestone and dolomite density-porosity. Furthermore, this study presents unsupervised artificial neural network (ANN) classification, based on five seismic attributes (instantaneous frequency, energy, band width, absorption quality factor, seismic amplitude), that comes in agreement with Ca–Mg ratio and the observed sonic transit time (DT log) variation with dolomitization/porosity increase. The hydrocarbon reservoir facies identified by the attributes classification explains the drilling results, with high accuracy/match to facies class centers, and can be used effectively in other settings. The integration, of multi-scale multi-data analysis and modeling, has provided a solid understanding of the reservoir-quality control and distribution. This study can be considered as a reliable platform for placing future infill wells in the study area, to lower the risk of drilling dry holes.

Classification of Design Attributes for FMCG (Fast Moving Consumer Goods) Products Official Store in E-commerce Website to Increase Usability and User Satisfaction

The purchase of FMCG products through e-commerce websites is currently growing rapidly and will continue to improve in the following years. E-commerce companies collaborating with the FMCG company to create an official store on e-commerce websites. Official store for food and personal care products is a strategy where e-commerce and FMCG company like create an official store where the FMCG company can directly sell their products officially with an online system on e-commerce websites. Official store on each e-commerce website has a different interface design that aims to increase user satisfaction and usability. This study aims to classify design parameters according to user expectations and satisfaction and give strategies to increase the usability of the official store in e-commerce websites for FMCG products. Result in this study indicated that attributes design provide maximum contribution to the overall usability and satisfaction of the websites.

Classification of key ecological attributes and stresses of biodiversity for ecosystem-based conservation assessments and management

A whole systems thinking approach to conservation has spawned new approaches in adaptive management planning that require a crucial understanding of what is essential for the functionality of ecosystems and the biodiversity they embrace. In this context, the key ecological attributes (KEA) have been introduced as aspects of a conservation target's biology or ecology that, if missing or altered, would lead to the loss of that target over time. Ecological stresses describe the impaired status of KEAs. Whilst for threats, the drivers of stresses, a systematic classification has been suggested and adopted by IUCN, all existing proposals for stresses and KEAs are preliminary. In order to fill the gap and provide conservation analysts and practitioners with a standard terminology supporting adaptive management planning we suggest a first hierarchical framework and comprehensive classification of key ecological attributes and corresponding stresses to biodiversity. Analyzing 22 vulnerability assessments in 13 countries, spread across 5 continents, as well as an extensive literature review, we identified 144 specific KEAs and stresses. These are differentiated and classified according to three hierarchical levels, 11 KEA and stress classes and 42 general KEAs and stresses. Our classification may help with describing and understanding both the natural functionality and also impaired functioning of biodiversity targets, as well as assist with the development of appropriate conservation strategies. The classification of key ecological attributes is presented as a list but it is important to recognize that the diverse array of KEAs and stresses are systemically interrelated across scales.

Improving User Attribute Classification with Text and Social Network Attention

User attribute classification is an important research topic in social media user profiling, which has great commercial value in modern advertisement systems. Existing research on user profiling has mostly focused on manually handcrafted features for different attribute classification tasks. However, these research has partly overlooked the social relation of users. We propose an end-to-end neural network model called the social convolution attention neural network. Our model leverages the convolution attention mechanism to automatically extract user features with respect to different attributes from social texts. The proposed model can capture the social relation of users by combining semantic context and social network information, and improve the performance of attribute classification. We evaluate our model in the gender, age, and geography classification tasks based on the dataset from SMP CUP 2016 competition, respectively. The experimental results demonstrate that the proposed model is effective in automatic user attribute classification with a particular focus on fine-grained user information. We propose an effective model based on the convolution attention mechanism and social relation information for user attribute classification. The model can significantly improve the accuracy in various user attribute classification tasks.

L’Habitat fortifié sur la frontière romaine présaharienne en Algérie orientale (Oasis des Ziban, Algérie)

Fortified settlement on the pre-Saharan Roman border in eastern Algeria between the 11th and 18th centuries. (Oasis of Ziban, Algeria). The fortified settlements on the foothills of the Zab Mountains occupy a geographical area which has long kept its status as a zone of cultural contact. In fact, after having belonged to the furthermost reaches of the Roman Empire, the pre-Saharan Zab made up the western boundary of Ifriqiyah with its location in the Gemellae sector in the south of Numidia . The intermingling of these two cultures inevitably led to the production of a Ziban landscape with its own characteristics. This article aims to go beyond a simple classification of the morphological attributes in the establishments situated on the left bank of the River Djedi, in the ancient confines of Numidia, to link the form of their dwellings with ancient Ziban lime structures, medieval principles of territorial division and the demands of irrigation. The cross-checking of texts concerning the historical geography of the Medieval and Ottoman eras, with material data from both excavations and the interpretation of cartographic records, together with the findings of on site surveys and investigations, has resulted in the formation of a corpus on these agglomerations in the western Zab Valley, while highlighting their fortified and hierarchical character. An analysis of the plans from historical records identified the morphological and functional attributes of all the establishments in this corpus of study, which resulted in their final classification into two groups : “ main” towns or “ secondary” agglomerations.

Using cluster analysis methods for multivariate mapping of traffic accidents

Abstract Many factors affect the occurrence of traffic accidents. The classification and mapping of the different attributes of the resulting accident are important for the prevention of accidents. Multivariate mapping is the visual exploration of multiple attributes using a map or data reduction technique. More than one attribute can be visually explored and symbolized using numerous statistical classification systems or data reduction techniques. In this sense, clustering analysis methods can be used for multivariate mapping. This study aims to compare the multivariate maps produced by the K-means method, K-medoids method, and Agglomerative and Divisive Hierarchical Clustering (AGNES) method, which among clustering analysis methods, with real data. The results from the study will suggest which clustering methods should be preferred in terms of multivariate mapping. The results show that the K-medoids method is more appropriate in terms of clustering success. Moreover, the aim is to reveal spatial similarities in traffic accidents according to the results of traffic accidents that occur in different years. For this aim, multivariate maps created from traffic accident data of two different years in Turkey are used. The methods are compared, and the use of the maps produced with these methods for risk management and planning is discussed. Analysis of the maps reveals significant similarities for both years.

A Sequential Process Model for Cognitive Diagnostic Assessment With Repeated Attempts.

When diagnostic assessments are administered to examinees, the mastery status of each examinee on a set of specified cognitive skills or attributes can be directly evaluated using cognitive diagnosis models (CDMs). Under certain circumstances, allowing the examinees to have at least one opportunity to correctly answer the questions and assessments, with repeated attempts on the items, provides many potential benefits. A sequential process model can be extended to model repeated attempts in diagnostic assessments. Two formulations of the sequential generalized deterministic-input noisy-"and"-gate (G-DINA) model were developed in this study. The first extension uses the latent transition analysis (LTA) approach to model changes in the attributes over attempts, and the second extension constructs a higher order structure of latent continuous variables and latent attributes to account for the dependences of the attributes over attempts. Accurate model parameter estimation and correct classifications of attributes were observed in a series of simulations using Bayesian estimation. The effectiveness of the developed sequential G-DINA model was demonstrated by fitting real data from a longitudinal mathematical test to the developed model and the longitudinal G-DINA model using the LTA approach. Finally, this article closes by discussing several important issues associated with the developed models and providing suggestions for future directions.

Computerized Adaptive Testing for Cognitively Based Multiple-Choice Data.

Cognitive diagnosis models (CDMs) are latent class models that hold great promise for providing diagnostic information about student knowledge profiles. The increasing use of computers in classrooms enhances the advantages of CDMs for more efficient diagnostic testing by using adaptive algorithms, referred to as cognitive diagnosis computerized adaptive testing (CD-CAT). When multiple-choice items are involved, CD-CAT can be further improved by using polytomous scoring (i.e., considering the specific options students choose), instead of dichotomous scoring (i.e., marking answers as either right or wrong). In this study, the authors propose and evaluate the performance of the Jensen-Shannon divergence (JSD) index as an item selection method for the multiple-choice deterministic inputs, noisy "and" gate (MC-DINA) model. Attribute classification accuracy and item usage are evaluated under different conditions of item quality and test termination rule. The proposed approach is compared with the random selection method and an approximate approach based on dichotomized responses. The results show that under the MC-DINA model, JSD improves the attribute classification accuracy significantly by considering the information from distractors, even with a very short test length. This result has important implications in practical classroom settings as it can allow for dramatically reduced testing times, thus resulting in more targeted learning opportunities.

QFD and Fuzzy Kano model based approach for classification of aesthetic attributes of SUV car profile

The aesthetic appearance and features of a product are the most censorious elements for the accomplishment of a product in the industry. An aesthetic is the quality element which adds value to the product design. Product design is a basic need of every manufacturing company in which visual aspects play an important role to enhance the customer satisfaction. Therefore, Quality Function Deployment (QFD) can be considered as an effective tool for translating the customer’s voice into the design of the product and its specifications. The Kano model helps to identify the desires of a product that brings greater satisfaction or dissatisfaction level to the customer. Kano model tells the connection between the product’s attributes and its satisfaction to the customer. For achieving better results, Fuzzy Kano model has been more favorably applied over traditional Kano model. In this work, an approach of Integration of Kano model into QFD has been applied with an aim to examine the customer satisfaction based on aesthetic sentiments. A Sport Utility Vehicle has been selected for the study. The aesthetic attributes have been selected with the help of QFD and their importance and classification have been calculated using both Fuzzy Kano and Traditional Kano model. The result of Fuzzy Kano and Traditional Kano model has also been compared to calculate the effectiveness of the applied approach.

Effectiveness of spectral data reduction in detection of salt-affected soils in a small study area

Data reduction is used to aggregate or amalgamate the large data sets into smaller and manageable information pieces in order to fast and accurate classification of different attributes. However, excessive spatial or spectral data reduction may result in losing or masking important radiometric information. Therefore, we conducted this research to evaluate the effectiveness of the different spectral data reduction algorithms including Principle Component Analysis (PCA) and Minimum Noise Fraction (MNF) transformation, Pixel Purity Index (PPI), and n Dimensional Visualizer (n-DV) algorithms on accuracy of the supervised classification of the salt-affected soils applying ETM+ data beside 188 ground control points. Results revealed that data reduction caused around 20 to 30 % decreases in classification results compared to none reduced data. It seems that applying spectral data reduction algorithm in small study areas is not only supportive, but also has negative effects on classification results. Therefore, it may better to not to use the algorithms in small areas.