Social discrimination (e.g., against females) arising from data mining techniques is a growing concern worldwide. In recent years, several methods have been proposed for making classifiers learned over discriminatory discrimination-aware. However, these suffer two major shortcomings: (1) They require either modifying the or tweaking specific classification algorithm and (2) are not flexible w.r.t. control multiple sensitive attribute handling. this paper, we present solutions...

Traffic incidents are nonrecurrent and pseudorandom events that disrupt the normal flow of traffic create a bottleneck in road network. The probability is higher during peak rates when systemwide effect most severe. Model-based solutions to incident detection problem have not produced practical, useful results primarily because complexity does lend itself accurate mathematical knowledge-based representations. A new multiparadigm intelligent system approach presented for solution problem,...

A general mathematical formulation is presented for the scheduling of construction projects and applied to problem highway scheduling. Repetitive nonrepetitive tasks, work continuity constraints, multiple-crew strategies, effects varying job conditions on performance a crew can be modeled. An optimization project problem, with goal minimizing direct cost. The nonlinear then solved by neural dynamics model developed recently Adeli Park. For any given duration, yields optimum schedule minimum...

In data mining we often have to learn from biased data, because, for instance, comes different batches or there was a gender racial bias in the collection of social data. some applications it may be necessary explicitly control this models This paper is first study learning linear regression under constraints that biasing effect given attribute such as batch number. We show how propensity modeling can used factoring out part justified by externally provided explanatory attributes. Then...

The task of automatic hate-speech and offensive language detection in social media content is utmost importance due to its implications unprejudiced society concerning race, gender, or religion. Existing research this area, however, mainly focused on the English language, limiting applicability particular demographics. Despite prevalence, Roman Urdu (RU) lacks resources, annotated datasets, models for task. In study, we: (1) Present a lexicon hateful words RU, (2) Develop an dataset called...

An adaptive computational model is presented for estimating the work zone capacity and queue length delay, taking into account following factors: number of lanes, open layout, length, lane width, percentage trucks, grade, speed, intensity, darkness factor, proximity ramps. The integrates judiciously mathematical rigor traffic flow theory with adaptability neural network analysis. A radial-basis function developed to learn mapping from quantifiable nonquantifiable factors describing control...

Automatic freeway incident detection is an important component of advanced transportation management systems (ATMS) that provides information for emergency relief and traffic control purposes. Earlier algorithms problems have produced less reliable results, especially in recurrent congestion compression wave conditions. This article presents a new two-stage single-station model based on wavelet analysis pattern recognition techniques. Wavelet used to denoise, cluster, enhance the raw data,...

A multiparadigm general methodology is advanced for development of reliable, efficient, and practical freeway incident detection algorithms. The performance the new fuzzy-wavelet radial basis function neural network (RBFNN) model Adeli Karim evaluated compared with benchmark California algorithm #8 using both real simulated data. evaluation based on three quantitative measures rate, false alarm time, qualitative measure portability. outperformed consistently under various scenarios. False...

A comprehensive evaluation is presented of the single-station wavelet energy neural network freeway incident-detection algorithm Karim and Adeli. Quantitative performance measures detection rate, false alarm time as well qualitative measure portability are investigated for both urban rural conditions. Further, compared with that California 8. This research demonstrates its excellent freeways across a wide range traffic flow roadway geometry conditions, regardless density loop detectors....

In a companion paper, an object-oriented (OO) information model was presented for construction scheduling, cost optimization, and change order management (CONSCOM), based on the creation of domain-specific development framework. The framework architecture is developed using generic software design elements, called patterns, which provide effective low-level solutions creating, organizing, maintaining objects. OO has been implemented in prototype system projects, CONSCOM, Microsoft Foundation...

An important advantage of cold-formed steel is the greater flexibility cross-sectional shapes and sizes available to structural designer. However, lack standard optimized makes selection most economical shape very difficult if not impossible. This task further complicated by complex highly nonlinear nature rules that govern their design. A general mathematical formulation computational model presented for optimization beams. The problem solved adapting robust neural dynamics Adeli Park,...

A case-based reasoning (CBR) model is presented for freeway work zone traffic management. The considers layout, demand, characteristics, control measures, and mobility impacts. four-set case base schema or domain theory developed to represent the cases based on aforementioned characteristics of problem. It includes a general information set, problem description solution (or control) an effects set. To improve interactivity CBR system its user-friendliness, hierarchical object-oriented...

Recently, the writers developed a general and powerful mathematical model for scheduling construction projects. An optimization formulation was presented with goal of minimizing direct cost. The nonlinear problem solved by recently patented neural dynamics Adeli Park. In this paper an object-oriented (OO) information is scheduling, cost optimization, change order management (CONSCOM) based on new model. to lay foundation generation flexible, powerful, maintainable, reusable software system...

We present an unsupervised method to find lexical variations in Roman Urdu informal text.Our includes a phonetic algorithm UrduPhone, featurebased similarity function, and clustering Lex-C.UrduPhone encodes roman strings their equivalent representations.This produces initial grouping of different spelling word.The function incorporates word features context.Lex-C is variant k-medoids that group variations.It threshold balance the number clusters maximum similarity.We test our system on two...