Adaptive Search System Research Articles

Model of the software and hardware complex of automated flight instruction for cadets based on the reference flight pattern

Information is provided regarding the developed model of the software-hardware complex of the automated flight instruction of cadets based on the reference image of the flight. It is proposed to model the movement of a helicopter between two support points as the movement of a material point that has mass and is affected by gravity. The movement model is represented by a system of differential equations of the first order, which describes the movement parameters: air speed, angle of inclination of the trajectory, flight height and distance of movement. Based on the existing regulatory rules for the operation of aircraft, a system of restrictions on movement parameters has been defined. The set of basic flight control bodies is substantiated, the laws of pilot control and restrictions on control parameters, which are also determined by the rules of helicopter operation, are set. The task of finding optimal control parameters at the stage of acceleration of a helicopter at a constant altitude has been formalized. The general structure of the adaptive search system has been built. Procedures for finding optimal parameters have been created using a combined method of finding optimal control parameters based on single-criteria conditional and vector optimization methods. Based on the model experiment, the optimal value of the parity criterion in the method of adaptive convolution of the optimal criterion vector is determined. The calculated data of the proposed model made it possible to find in real time the parameters of the movement of the helicopter under the control of the pilot, the history of the deviation of this movement from the ideal, and the dynamic forecast of such deviation. The possibility of obtaining predicted solutions in the case of existing linear control laws or approximate to linear control laws has been established experimentally. The parameters of the model were determined experimentally. A system of commands of an electronic instructor in verbal form is proposed for correcting the pilot's behavior and bringing his actions closer to optimal ones. Recommendations regarding the practical application of the developed model are substantiated.

Extracting the Implicit Search States from Explicit Behavioral Signals in Complex Search Tasks

AbstractIn a complex search task, users often go through different search states during search interactions under complex search tasks. Automatically identify these search states from observable behavioral measures will enhance our understanding of users' cognitive variations and serve as the basis for state‐aware adaptive search recommendations and evaluations. To achieve this goal, this state seeks to extract implicit search states from explicit behavioral signals and explore the possible connection between automatically clustered search states and the existing state typologies developed based on user labeling and expert annotations. The results from our preliminary work indicate that search states and the associated behavioral patterns can be extracted through clustering analysis in multiple datasets, and that the identified clusters/states can be mapped to the states identified via qualitative coding. Our study demonstrates the feasibility of automatically extracting search states and pave the path towards state‐aware adaptive search systems.

An Improved Grasshopper Optimizer for Global Tasks

The grasshopper optimization algorithm (GOA) is a metaheuristic algorithm that mathematically models and simulates the behavior of the grasshopper swarm. Based on its flexible, adaptive search system, the innovative algorithm has an excellent potential to resolve optimization problems. This paper introduces an enhanced GOA, which overcomes the deficiencies in convergence speed and precision of the initial GOA. The improved algorithm is named MOLGOA, which combines various optimization strategies. Firstly, a probabilistic mutation mechanism is introduced into the basic GOA, which makes full use of the strong searchability of Cauchy mutation and the diversity of genetic mutation. Then, the effective factors of grasshopper swarm are strengthened by an orthogonal learning mechanism to improve the convergence speed of the algorithm. Moreover, the application of probability in this paper greatly balances the advantages of each strategy and improves the comprehensive ability of the original GOA. Note that several representative benchmark functions are used to evaluate and validate the proposed MOLGOA. Experimental results demonstrate the superiority of MOLGOA over other well-known methods both on the unconstrained problems and constrained engineering design problems.

Multitasking, cognitive coordination and cognitive shifts during web searching

Enlightened by the cognitive psychology literature demonstrating that multitasking behavior is correlated closely to humans' cognitive processing and coordinationcapabilities, this doctoral thesis explores the nature and relationship of multitasking, cognitive coordination, and cognitive shifts during Web searching. A combination of data collection techniques including pre- and post-questionnaires, think-aloud protocols, search logs, observations and semi-structured interviews were employed to observe how 42 postgraduate students cognitively coordinated Web searches when multitasking across 126 real-life information problems. Results show that Web searching is a dynamic interaction between users and Web search systems during which multiple information problems ordering, evolving information problems generating, searching task switching, task and mental coordinating occur, and at a deeper level, cognitive shifts take place. The explicit task level coordination is closely linked to multitasking behavior. The implicit cognitive level coordination is related to the task coordination process; this includes the evolution of information problem development and search task switching. Coordination mechanisms directly result in the shifts of cognitive states including strategy, evaluation, and view states which further affect users' holistic shifts in information problem understanding and knowledge contribution. Based on the findings, a Web search model integrating multitasking, cognitive coordination and cognitive shifts (MCC model) is presented. Cognitive coordination as humans' vital capability under conscious control plays a nexus role in supporting multitasking Web searches and invoking the occurrence of shifts in cognition. The results extend multitasking research and Web search models in the field by examining and integrating underlying cognitive and coordination support mechanisms. Design recommendations for adaptive search systems are also put forward based on the characteristics and relationship of users' behaviors in multitasking, cognitive coordination and shifting.

Predicting stock market trends using hybrid ant-colony-based data mining algorithms: an empirical validation on the Bombay Stock Exchange

Ant Colony Optimisation (ACO) algorithms use simple mutually cooperating agents (ants) to produce a robust and adaptive search system, which can be used for knowledge discovery. In this paper, a Support Vector Machine (SVM)-cAnt-Miner-based system for predicting the next-day's trend in stock markets is proposed. The trend predicted by the proposed system is then used to identify the appropriate time to buy and sell securities. Performance of the proposed system is evaluated against SVM-Ant-Miner, SVM-Ant-Miner2, Naive-Bayes and an Artificial Neural Network (ANN)-based trend prediction system. The results indicate that the proposed system outperforms all the other techniques considered.

Guest editorial: Machine learning in and for music

Music has always held a special fascination for researchers in mathematics and computer science. It is both an abstract, formal system and a physical event; it relates to areas of human endeavour as diverse as the arts, aesthetics, philosophy, mathematics, acoustics, physiology, or history; and it poses deep questions regarding human nature, intellect, and emotions. Artificial Intelligence, in particular, has dreamed of devising intelligent musical machines almost from its inception. Music has also played its role in the history of machine learning research. Initially, musical tasks were mostly used as toy domains to demonstrate the properties of new learning algorithms, and the musical problems addressed were usually greatly simplified (the guest editor’s own early research is no exception). But recent years have seen a dramatic change in this regard. Machine learning researchers have begun to address musical tasks of real-world complexity—from the analysis of the artistic style of famous musicians to systems that learn to accompany, or improvise with, human musicians, and from programs that classify audio recordings into musical genres to computers that can recognise artists from the way they play. This trend towards ‘real-world’ applications has been accelerated by the dramatic changes that are currently going on in the digital music market. With the Internet becoming the central medium for the distribution of music and Web-based music stores offering literally millions of digital music tracks, there is a pressing need for things like intelligent and adaptive music search and recommendation systems, automatic style recognition and music classification, etc. The new field of Music Information Retrieval (MIR) is a direct response to this need, and the research published at the International Conference on Music Information Retrieval (ISMIR) and similar events is closely watched by major players in the music industry. The potential of this kind of research is also being recognised by the large international research funding organisations. At the European level, for instance, several large-scale research efforts in the area of intelligent computing and learning in music and audio have recently been funded—e.g., the projects SIMAC (www.semanticaudio.org), Semantic HiFi (http://shf.ircam.fr), and S2S2 (http://www.s2s2.org).

The conference call search problem in wireless networks

Cellular telephony systems, where locations of mobile users are unknown at some times, are becoming more common. In such systems, mobile users are roaming in a zone and a user reports its location only if it leaves the zone entirely. The conference call search (CCS) problem deals with tracking a set of mobile users in order to establish a call. To find a single roaming user, the system may need to search each cell where the user may be located. The goal is to identify the location of all users, within bounded time, satisfying some additional constraints on the search scheme. We consider cellular systems with n cells and m mobile users (cellular phones). The uncertain location of users is given by m probability distribution vectors. Whenever the system needs to find the users, it conducts a search operation lasting at most d rounds. A request for a single search step specifies a user and a cell. In this search step, the cell is asked whether the given user is located there. In each round the system may perform an arbitrary number of such requests. An integer number B ⩾ 1 bounds the number of distinct requests per cell in every round. The bound d results from quality of service considerations, whereas the bound B results from the wireless bandwidth allocated for signaling being scarce. Every search step consumes expensive wireless links, which motivates search techniques minimizing the expected number of requests thus reducing the total search costs. We distinguish between oblivious, semi-adaptive and adaptive search protocols. An oblivious search protocol decides on all requests in advance, and stops only when all users are found. A semi-adaptive search protocol decides on all the requests in advance, but it stops searching for a user once it is found. An adaptive search protocol stops searching for a user once it has been found (and its search strategy may depend on the subsets of users that were found in each previous round). We establish the difference between those three search models. We show that for oblivious “single query per cell” systems ( B = 1 ), and a tight environment ( d = m ), it is NP-hard to compute an optimal solution (the case d = m = 2 was proven to be NP-hard already by Bar-Noy and Naor) and we develop a PTAS for these cases (for fixed values of d = m ). However, we show that semi-adaptive systems allow polynomial time algorithms. This last result also shows that the case B = 1 and d = m = 2 is polynomially solvable also for adaptive search systems, answering an open question of Bar-Noy and Naor.

Adaptive Control System of the Arc Welding Equipment

There is presented investigation into defects of the arc welding flanges, they reasons and factors, which decide on quality of formation weld seams. If you like to optimize welding process, it is necessary to stabilize not only the speed of electric drives of the filler wire and little cart, but also the arc current, voltage, lengh and electrode wire melting speed. The formation principles of the adaptive control system there are presented, the function and structural schemes are obtained. Research results shows, that by employment of technical vision system and mathematical model of weld seams describing with logical operators the adaptive search system may be created. This system may quickly to change its parametres and structure, and in that way to optimize the same welding process. The problem on synthesis of the welding control system with the changeable structure and parameters was solved by using computer package Kvazio 1 with simplex search algorithms. The change laws of the amplification coefficients of the proportional and differential regulators where obtained. Modelling results shows, that the transitional process of the arc voltage is aperiodiocal and intimate to optimal. Because that is stabilizing the arc current, arc buring process and geometrical form of the weld seam. Ill. 6, bibl. 7 (in Lithuanian; summaries in Lithuanian, English and Russian).

The design of an adaptive agent driven search system for empowering disenfranchised users

This research is about designing and testing a usable, adaptive search system for disenfranchised users. It involves developing the adaptive agent model and architecture for the tool as well as researching its effectiveness for the users. The potential users for this system are not only naïve about computer driven technology but are also inhibited to try one, which is a consequence of digital divide. They have not experienced a technology-driven environment where computer has become just another tool to mechanize the daily lives. In this respect they are disenfranchised and they belong to the segment that falls under Poor America. Naturally, handing over another search system will not bridge the divide or empower the disenfranchised users. In order to help them overcome the inhibition, this tool has to be, at the very least, easy to use. Therefore as part of the solution, an adaptive, agent-driven search system is being developed. The system will adapt and personalize the process of search to support individual goals and social needs.

Recognizing patterns in information retrieval: a memory-based classifier for inferring relevancy

An application of memory-based reasoning (MBR) for determining the relevancy of retrieved information, customized for the needs and preferences of an individual user, is presented. The use of MBR method in conjunction with a classification scheme such as k-nearest neighbors ( k-NN) can set the foundation for an intelligent agent for search of patterns in databases. An intelligent agent is an adaptive database search system that can help reduce user information overload problem. To this end, a system called VINAYAK, † † Vinayak is one of the names of Lord Ganesha, who represents wisdom and knowledge in Hindu mythology. has been developed to investigate various issues concerned with information retrieval as an automated function. Current experiments have focused on chemical engineering literature search and safety-related retrieval through nuclear databases. The results point towards the advantage of the use of intelligent agent methodology in conjunction with a database search engine.