Explainable Rules and Heuristics in AI Algorithm Recommendation Approaches——A Systematic Literature Review and Mapping Study

The exponential use of artificial intelligence(AI)to solve and automated complex tasks has catapulted its popularity generating some challenges that need to be addressed.While AI is a powerfulmeans to discover interesting patterns and obtain predictive models,the use of these algorithms comes with a great responsibility,as an incomplete or unbalanced set of training data or an unproper interpretation of the models’outcomes could result in misleading conclusions that ultimately could become very dangerous.For these reasons,it is important to rely on expert knowledge when applying these methods.However,not every user can count on this specific expertise;non-AIexpert users could also benefit from applying these powerful algorithms to their domain problems,but they need basic guidelines to obtain themost out of AI models.The goal of this work is to present a systematic review of the literature to analyze studies whose outcomes are explainable rules and heuristics to select suitable AI algorithms given a set of input features.The systematic review follows the methodology proposed by Kitchenham and other authors in the field of software engineering.As a result,9 papers that tackle AI algorithmrecommendation through tangible and traceable rules and heuristics were collected.The reduced number of retrieved papers suggests a lack of reporting explicit rules and heuristics when testing the suitability and performance of AI algorithms.

Exact and heuristic formulae to compute the geodetic height from the ellipse equation

The conversion of the cartesian coordinates of a point to its geodetic equivalent coordinates in reference to the geodetic ellipsoid is one of the main challenges in geodesy.The ellipse equation in the meridian plane significantly influences the value of the geodetic coordinates.This research analyzes this influence and how it can contribute to their solutions.The study investigates the mathematical relation between them and presents an exact formula relating to the geodetic height and the ellipse equation.In addition,a heuristic formula for the relation between the geodetic height and the ellipse equation is proposed,which is independent of the geodetic latitude and has a relative accuracy better than 99.9 %.The calculation is stable,and the cost is low.

Improved STNModels and Heuristic Rules for Cooperative Scheduling in Automated Container Terminals

Improving the cooperative scheduling efficiency of equipment is the key for automated container terminals to copewith the development trend of large-scale ships. In order to improve the solution efficiency of the existing spacetimenetwork (STN) model for the cooperative scheduling problem of yard cranes (YCs) and automated guidedvehicles (AGVs) and extend its application scenarios, two improved STN models are proposed. The flow balanceconstraints in the original model are decomposed, and the trajectory constraints of YCs and AGVs are added toacquire the model STN_A. The coupling constraint in STN_A is updated, and buffer constraints are added toSTN_A so that themodel STN_B is built.As the size of the problem increases, the solution speed of CPLEX becomesthe bottleneck. So a heuristic method containing three groups of heuristic rules is designed to obtain a near-optimalsolution quickly. Experimental results showthat the computation time of STN_A is shortened by 49.47% on averageand the gap is reduced by 1.69% on average compared with the original model. The gap between the solution ofthe heuristic rules and the solution of CPLEX is less than 3.50%, and the solution time of the heuristic rules is onaverage 99.85% less than the solution time of CPLEX. Compared with STN_A, the computation time for solvingSTN_B increases by 58.93% on average.

Estimating regional coal resource efficiency in China using three-stage DEA and bootstrap DEA models

Based on three-stage data envelopment analysis(DEA) model estimates of resource utilization efficiency of coal,we selected 29 provinces from China's 2012 input-output data and used the bootstrap DEA model to correct the bias.The results show that the mean overall technical efficiency,pure technical efficiency,and scale efficiency was 0.833,0.998,and 0.711 in 2012,respectively.Moreover,the comprehensive technical efficiency score indicates that the scale is invalid.Area utilization efficiencies for the eastern,central,and western regions were 0.917,0.731,and 0.629,respectively.The results prove that there are significant differences in the distribution of coal resources utilization efficiency across regions.

Investigation on seismic response of a three-stage soil slope supported by anchor frame structure

Based on a typical prototype of a soil slope in engineering practice, a numerical model of a three-stage soil slope supported by the anchor frame structure was established by means of FLAC3D code. The dynamic responses of three-stage soil slope and frame structure were studied by performing a series of bidirectional Wenchuan motions in terms of the failure mode of three-stage structure, the acceleration of soil slope, the displacement of frame structure, and the anchor stress of frame structure. The response accelerations in both horizontal and vertical directions are the most largely amplified at the slope top of each stage subjected to different shaking cases. The platforms among the stages reduce the amplification effect of response acceleration. The residual displacement of frame structure increases significantly as the intensity of shaking case increases. The frame structure at each stage presents a combined displacement mode consisting of a translation and a rotation around the vertex. The anchor stress of frame structure is mainly increased by the first intense pulse of Wenchuan seismic wave, and it is sensitive to the intensity of shaking case. The anchor stress of frame structure at the first stage is the most considerably enlarged by earthquake loading.

Production Efficiency of Two-Oriented Agriculture Based on Three-Stage DEA Model

Two-oriented agriculture was a complex organism coupling production,economics,society and ecology.Its development process was affected by various factors such as producers,nature,society,etc.In order to overcome measurement error of traditional data envelopment analysis caused by ignoring random,three-stage DEA model was studied to remove environmental factors and random effects.On the foundation of this model was two-oriented agriculture comprehensive production efficiency of 14 cities were estimated in Hunan Province in 2008,and brown forth corresponding policy proposals to promote agricultural development.

Analyzing and De-Anonymizing Bitcoin Networks:An IP Matching Method with Clustering and Heuristics

The anonymity and de-anonymity of blockchain and Bitcoin have always been a hot topic in blockchain related research.Since Bitcoin was created by Nakamoto in 2009,it has,to some extent,deviated from its currency attribute as a trading medium but instead turned into an object for financial investment and operations.In this paper,the power-law distribution that the Bitcoin network obeys is given with mathematical proof,while traditional deanonymous methods such as clustering fail to satisfy it.Therefore,considering the profit-oriented characteristics of Bitcoin traders in such occasion,we put forward a de-anonymous heuristic approach that recognizes and analyzes the behavioral patterns of financial High-Frequency Transactions(HFT),with realtime exchange rate of Bitcoin involved.With heuristic approach used for de-anonymity,algorithm that deals with the adjacency matrix and transition probability matrix are also put forward,which then makes it possible to apply clustering to the IP matching method.Basing on the heuristic approach and additional algorithm for clustering,finally we established the de-anonymous method that matches the activity information of the IP with the transaction records in blockchain.Experiments on IP matching method are applied to the actual data.It turns out that similar behavioral pattern between IP and transaction records are shown,which indicates the superiority of IP matching method.

Layer-layout-based heuristics for loading homogeneous items into a single container

The container loading problem (CLP) is a well-known NP-hard problem. Due to the computation complexity, heuristics is an often-sought approach. This article proposes two heuristics to pack homogeneous rectangular boxes into a single container. Both algorithms adopt the concept of building layers on one face of the container, but the first heuristic determines the layer face once for all, while the second treats the remaining container space as a reduced-sized container after one layer is loaded and, hence, selects the layer face dynamically. To handle the layout design problem at a layer's level, a block-based 2D packing procedure is also developed. Numerical studies demonstrate the efficiency of the heuristics.

A Hybrid Heuristics for Irregular Flight Recovery

Adverse weather conditions,congestion at airports,and mechanical failures often disrupt regular flight schedules. The irregular flight recovery problem aims to recover these schedules through reassignments of flights and cancellations. In this article,we develop the classic resource assignment model for the irregular flight recovery problem,and a new hybrid heuristic procedure based on greedy random adaptive search procedure （GRASP） and simulated annealing algorithm is presented to solve this problem. As compared with the original GRASP method,the proposed algorithm demonstrates quite a high global optimization capability. Computational experiments on large-scale problems show that the proposed procedure is able to generate feasible revised flight schedules of good quality in less than five seconds.

Research on Different Heuristics for Minimax Algorithm Insight from Connect-4 Game

Minimax algorithm and machine learning technologies have been studied for decades to reach an ideal optimization in game areas such as chess and backgammon. In these fields, several generations try to optimize the code for pruning and effectiveness of evaluation function. Thus, there are well-armed algorithms to deal with various sophisticated situations in gaming occasion. However, as a traditional zero-sum game, Connect-4 receives less attention compared with the other members of its zero-sum family using traditional minimax algorithm. In recent years, new generation of heuristics is created to address this problem based on research conclusions, expertise and gaming experiences. However, this paper mainly introduced a self-developed heuristics supported by well-demonstrated result from researches and our own experiences which fighting against the available version of Connect-4 system online. While most previous works focused on winning algorithms and knowledge based approaches, we complement these works with analysis of heuristics. We have conducted three experiments on the relationship among functionality, depth of searching and number of features and doing contrastive test with sample online. Different from the sample based on summarized experience and generalized features, our heuristics have a basic concentration on detailed connection between pieces on board. By analysing the winning percentages when our version fights against the online sample with different searching depths, we find that our heuristics with minimax algorithm is perfect on the early stages of the zero-sum game playing. Because some nodes in the game tree have no influence on the final decision of minimax algorithm, we use alpha-beta pruning to decrease the number of meaningless node which greatly increases the minimax efficiency. During the contrastive experiment with the online sample, this paper also verifies basic characters of the minimax algorithm including depths and quantity of features. According to the experiment, these two characters can both effect the decision for each step and none of them can be absolutely in charge. Besides, we also explore some potential future issues in Connect-4 game optimization such as precise adjustment on heuristic values and inefficiency pruning on the search tree.

An Analysis of Operational Efficiency of Agricultural Insurance Agency:Based on Three-Stage DEA Model

From the perspective of agricultural insurance agency,this paper focuses on the study of operational efficiency of agricultural insurance agency,and analyzes the operational efficiency of agricultural insurance agency and its influencing factors,in order to improve the operational efficiency of agricultural insurance agency,better supply agricultural insurance and achieve the policy effect of agricultural insurance.The results of this paper are as follows:( i) Either comprehensive operational efficiency or scale efficiency of Chinese-funded agency is higher than that of foreign-funded agency,but its pure technical efficiency is lower than that of foreign-funded agency,indicating that the managerial decision ability of Chinese-funded agency is weaker than that of foreign-funded agency,and needs to be improved;( ii) The operational efficiency of professional agricultural insurance agency is higher than that of comprehensive agricultural insurance agency,and the agricultural insurance agency is greatly affected by environmental factors;( iii) The operating time of agricultural insurance agency is proportional to its operational efficiency;( iv) The quality of employees is positively correlated with the capital and cost input difference,but negatively correlated with the difference in the number of employees. Therefore,it is necessary to pay attention to the improvement of employees' working efficiency while laying emphasis on employees' quality.

Contribution to the Optimization of the Transient Stability of an Electric Power Transmission Network Using a Universal Power Flow Compensator Controlled by a Three-Stage Inverter

The use of an electrical network as close as possible to its limits can lead to its instability in the event of a high amplitude disturbance. The damping of system oscillations can be achieved by conventional means of voltage and speed regulation but also by FACTS (Flexible AC Transmission Systems) devices, which are increasingly used in power networks. In this work, optimal control coordination between a hybrid power flow controller and a three-level inverter was used to improve the transient stability of a transmission line. The UPFC is a combination of a serial compensator (SSSC) and a parallel compensator (STATCOM) both connected to a DC-LINK DC bus. The SSSC acts as a voltage source for the network and injects a voltage that can be adjusted in phase and amplitude in addition to the network voltage;the STATCOM acts as a current source. The approach used is tested in the Matlab Simulink environment on a single machine network. Optimal controller tuning gives a better transient stability improvement by reducing the transport angle oscillations from 248.17% to 9.85%.

A new heuristics model of simulating pedestrian dynamics based on Voronoi diagram

A new heuristics model based on the Voronoi diagram is presented to simulate pedestrian dynamics with the noncrowded state, in which these mechanisms of preference demand evading and surpassing, microscopic anti-deadlock, and site-fine-tuning are considered. The preference demand describes the willingness determination of detouring or following other pedestrians. In the evading and surpassing mechanisms, in order to achieve a balance between avoiding conflicts and minimizing detour distances, a new pair of concepts: "allow-areas and denial-areas" are introduced to divide the feasible region for pedestrians detour behaviors, in which the direction and magnitude of detour velocity are determined.A microscopic anti-deadlock mechanism is inserted to avoid deadlock problem of the counter-directional pedestrian. A site-fine-tuning mechanism is introduced to describe the behavior of avoiding getting too close to the neighbors in pedestrian movement. The presented model is verified through multiple scenarios, including the uni-or bi-direction pedestrian flow in the corridor without obstacles, the uni-direction pedestrian flow in the corridor with obstacles, and the pedestrian evacuation from a room with single-exit. The simulation results show that the velocity–density relationship is consistent with empirical data. Some self-organizing phenomena, such as lanes formation and arching are observed in the simulation.When pedestrians detour an obstacle, the avoiding area before the obstacle and the unoccupied area after the obstacle can be observed. When pedestrians evacuate through a bottleneck without panic, the fan-shaped crowd can be found, which is consistent with the actual observation. It is also found that the behavior of following others in an orderly manner is more conducive to the improvement of the overall movement efficiency when the crowd moves in a limited space.

Improving Bandwidth of Three-Stage Electro-Hydraulic Servo Valve Based on Speed-Feedback

In order to improve the dynamic response bandwidth of three-stage electro-hydraulic servo valve,a new method,speed-feedback control is presented in this paper.The construction and principle of three-stage electro-hydraulic servo valve are explained,and the mathematical model of three-stage electro-hydraulic servo valve is built in frequency domain.Experimental and simulation results show that the bandwidth compared with proportional control is improved under speed-feedback control.Moreover,the research results play an important role in developing high performance three-stage electro-hydraulic servo valve.

Numerical Simulation of Enhanced Oil-Water Separation in a Three-Stage Double-Stirring Extraction Tank

Numerical simulation of enhanced fluid flow characteristics in a three-stage double-stirring extraction tank was conducted with the coupling of an Eulerian multiphase flow model and a Morsi-Alexander interphase drag force model. Results show that the addition of a stirring device into the settler can efficiently reduce the volume fraction of out-of-phase impurity in the outlet, and accelerate the settling separation of oil-water mixture. Such addition can also effectively break down the oil-water-wrapped liquid droplets coming from the mixer, inhibit reflux from the outlet, and improve the oil-water separation. The addition of a stirring device induces ignorable power consumption compared with that by the mixer, and can thus facilitate the commercialized promotion of this novel equipment.

How Group Heuristics Promotes Cooperative Behavior in Social Dilemma SituationmExploration of Group Heuristics Effects on the Basis of Computer Simulation

As it is impossible to assume complete rationality in a social dilemma situation, the assumption of bounded rationality is appropriate. Under the bounded rationality, it would be reasonable to assume that one behaves according to the heuristics principle. The group identity effect in a social dilemma situation might be very important in order to attain cooperation. The aim of this study was to clarify how to promote a cooperative behavior by avoiding a social dilemma situation. The group heuristics was taken into account, and it was explored how the group heuristics promotes a cooperative behavior in a social dilemma situation. As a result of a two-person game theory experiment, the group heuristics was found to play an important role in a social dilemma situation, and enhance a cooperative behavior. For the following three cases, the higher cooperation rate was attained at the latter half of the experiment： （a） mutual in-group condition, （b） one-way in-group condition, and （f） one-way unknown condition （in-group）. In conclusion, the consciousness of in-group membership might help to promote actively mutual cooperation.

Flower Pollination Heuristics for Nonlinear Active Noise Control Systems

In this paper,a novel design of the flower pollination algorithm is presented for model identification problems in nonlinear active noise control systems.The recently introduced flower pollination based heuristics is implemented to minimize the mean squared error based merit/cost function representing the scenarios of active noise control system with linear/nonlinear and primary/secondary paths based on the sinusoidal signal,random and complex random signals as noise interferences.The flower pollination heuristics based active noise controllers are formulated through exploitation of nonlinear filtering with Volterra series.The comparative study on statistical observations in terms of accuracy,convergence and complexity measures demonstrates that the proposed meta-heuristic of flower pollination algorithm is reliable,accurate,stable as well as robust for active noise control system.The accuracy of the proposed nature inspired computing of flower pollination is in good agreement with the state of the art counterpart solvers based on variants of genetic algorithms,particle swarm optimization,backtracking search optimization algorithm,fireworks optimization algorithm along with their memetic combination with local search methodologies.Moreover,the central tendency and variation based statistical indices further validate the consistency and reliability of the proposed scheme mimic the mathematical model for the process of flower pollination systems.

A Heuristics-Based Cost Model for Scientic Workow Scheduling in Clou

Scientic Workow Applications(SWFAs)can deliver collaborative tools useful to researchers in executing large and complex scientic processes.Particularly,Scientic Workow Scheduling(SWFS)accelerates the computational procedures between the available computational resources and the dependent workow jobs based on the researchers’requirements.However,cost optimization is one of the SWFS challenges in handling massive and complicated tasks and requires determining an approximate(near-optimal)solution within polynomial computational time.Motivated by this,current work proposes a novel SWFS cost optimization model effective in solving this challenge.The proposed model contains three main stages:(i)scientic workow application,(ii)targeted computational environment,and(iii)cost optimization criteria.The model has been used to optimize completion time(makespan)and overall computational cost of SWFS in cloud computing for all considered scenarios in this research context.This will ultimately reduce the cost for service consumers.At the same time,reducing the cost has a positive impact on the protability of service providers towards utilizing all computational resources to achieve a competitive advantage over other cloud service providers.To evaluate the effectiveness of this proposed model,an empirical comparison was conducted by employing three core types of heuristic approaches,including Single-based(i.e.,Genetic Algorithm(GA),Particle Swarm Optimization(PSO),and Invasive Weed Optimization(IWO)),Hybrid-based(i.e.,Hybrid-based Heuristics Algorithms(HIWO)),and Hyper-based(i.e.,Dynamic Hyper-Heuristic Algorithm(DHHA)).Additionally,a simulation-based implementation was used for SIPHT SWFA by considering three different sizes of datasets.The proposed model provides an efcient platform to optimally schedule workow tasks by handing data-intensiveness and computational-intensiveness of SWFAs.The results reveal that the proposed cost optimization model attained an optimal Job completion time(makespan)and total computational cost for small and large sizes of the considered dataset.In contrast,hybrid and hyper-based approaches consistently achieved better results for the medium-sized dataset.

A Unique Orogenic System with Three-Stage Rapakivi Granites in the World：Proterozoic,Palaeozoic and Mesozoic Rapakivi Granites in the China Central Orogenic System

Objective Rapakivi granites,characterized by rapakivi texture,Atype granite feature and an anorogenic setting,commonly occur in the Proterozoic of the Northern Hemisphere（Fig.la）.Recently,more and more Phanerozoic rapakivi granite suites have been identified and some even occur in orogenic belts.Significantly,three-stage,Proterozoic.

A Three-section Algorithm of Dynamic Programming Based on Three-stage Decomposition System Model for Grade Transition Trajectory Optimization Problems

This paper introduces a practical solving scheme of gradetransition trajectory optimization(GTTO) problems under typical certificate-checking–updating framework. Due to complicated kinetics of polymerization,differential/algebraic equations(DAEs) always cause great computational burden and system non-linearity usually makes GTTO non-convex bearing multiple optima. Therefore, coupled with the three-stage decomposition model, a three-section algorithm of dynamic programming(TSDP) is proposed based on the general iteration mechanism of iterative programming(IDP) and incorporated with adaptivegrid allocation scheme and heuristic modifications. The algorithm iteratively performs dynamic programming with heuristic modifications under constant calculation loads and adaptively allocates the valued computational resources to the regions that can further improve the optimality under the guidance of local error estimates. TSDP is finally compared with IDP and interior point method(IP) to verify its efficiency of computation.