- TAKING THE FIFTH Brendan Emmett Quigley (www.brendanemmettquigley.com) 1 14 17 20 23 33 37 40 43 50 55 60 63 2 34 51 3 28 52 4 24 46 21 44 61 64 5 15 18 38 41 56 6 35 57 7 25 29 58 8 26 47 53 9 27 45 10 22 39 42 19 36 59 62 65 11 16 30 54 12 31 48 13 32 49 ©2020 ACROSS 1. Fifth Avenue 5. Mexican turnip-like tuber 11. Years on the planet 14.
- Every leader needs to take to succeed. I loved this book twenty-five years ago, and I love it today.” —Joel Kurtzman, author, Common Purpose; editor-in-chief, Korn/Ferry Institute’s Briefings on Talent & Leadership “We consider this twenty-fifth anniversary, fifth edition, the best leadership book.
- Updated: Dec 16th, 2020. “Taking the Fifth” refers to a person’s invocation of his or her Fifth Amendment right under the U.S. Constitution to refuse to give statements that could be used against the speaker in a criminal proceeding. The right attaches when people are questioned by the police (it’s the basis for the Miranda warning s.
- Due to a planned power outage, our services will be reduced today (June 15) starting at 8:30am PDT until the work is complete. We apologize for the inconvenience.
‘Taking the Fifth’ in Bankruptcy. Kippes and Erin Marie Schmidt. Trial Attorneys, Office of the United States Trustee, Dallas. The Fifth Amendment of the Constitution encompasses many important rights granted to criminal defendants – including the safeguard against double jeopardy and the right to due.
Contents lists available at SciVerse ScienceDirect
Knowledge-Based Systems journal homepage: www.elsevier.com/locate/knosys
A new Fruit Fly Optimization Algorithm: Taking the ﬁnancial distress model as an example Wen-Tsao Pan ⇑ Department of Information Management, Oriental Institute of Technology, Taiwan, ROC
a r t i c l e
i n f o
Article history: Received 29 May 2011 Received in revised form 29 June 2011 Accepted 1 July 2011 Available online 8 July 2011 Keywords: Fruit Fly Optimization Algorithm Financial distress Optimization problem General Regression Neural Network Data mining
a b s t r a c t The treatment of an optimization problem is a problem that is commonly researched and discussed by scholars from all kinds of ﬁelds. If the problem cannot be optimized in dealing with things, usually lots of human power and capital will be wasted, and in the worst case, it could lead to failure and wasted efforts. Therefore, in this article, a much simpler and more robust optimization algorithm compared with the complicated optimization method proposed by past scholars is proposed; the Fruit Fly Optimization Algorithm. In this article, throughout the process of ﬁnding the maximal value and minimal value of a function, the function of this algorithm is tested repeatedly, in the mean time, the population size and characteristic is also investigated. Moreover, the ﬁnancial distress data of Taiwan’s enterprise is further collected, and the fruit ﬂy algorithm optimized General Regression Neural Network, General Regression Neural Network and Multiple Regression are adopted to construct a ﬁnancial distress model. It is found in this article that the RMSE value of the Fruit Fly Optimization Algorithm optimized General Regression Neural Network model has a very good convergence, and the model also has a very good classiﬁcation and prediction capability. Ó 2011 Elsevier B.V. All rights reserved.
1. Preface In recent years, the treatment of optimization problems has caught everyone’s attention, for example, the production schedule optimization problem [1,9], the shortest route problem in the logistics industry [13,12] or the shift arrangement problem in the trafﬁc transport industry [7,22], etc. should all be treated by the optimization algorithm. Until now, algorithms that are commonly used to treat optimization problems and data mining include: regression and classiﬁcation [18–21], genetic algorithm [14,2], Ant Colony Optimization Algorithm [6,4] and Particle Swarm Optimization Algorithm [8,25]. However, the common disadvantages of these algorithms are complicated computational processes, difﬁculty of understanding for beginners and there are very many parameters in GA-base methods. Therefore, this article author Pan  proposes a new optimization algorithm, which is called the Fruit Fly Optimization Algorithm or Fly Optimization Algorithm (abbreviated as FOA). Such an optimization algorithm has advantages such as a simple computational process, ease of transformation of such concept into program code and ease of understanding, etc. In this article, the way of ﬁnding maximal value and minimal value of function is ﬁrst used to test repeatedly, the function of this optimization algorithm. In the meantime, the correlation between the population size and ⇑ Address: 3F., No.12, Lane 271, Longjiang Rd., Jhongshan District, Taipei City 104, Taiwan, ROC. E-mail address: [email protected] 0950-7051/$ - see front matter Ó 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.knosys.2011.07.001
the optimization capability of the fruit ﬂy group is investigated. Furthermore, this article refer to Ravisankar , Li , Cho  and Hadavandi  adopts the ﬁnancial distress data of enterprises with stocks listed in the regular stock market and overthe-counter stock market from the years 2003 to 2004 as the test data. Meanwhile, fruit ﬂy algorithm optimized General Regression Neural Network (abbreviated as FOAGRNN), General Regression Neural Network (abbreviated as GRNN) and Multiple Regression (abbreviated as MR) are used to set up a ﬁnancial distress model. Finally, the classiﬁcation prediction capabilities of these three models are compared to verify the feasibility of applying the Fruit Fly Optimization Algorithm in real cases. The main structure of this article is as follows: The ﬁrst section introduces the research motivation and objective of this article. The second section introduces the Fruit Fly Optimization Algorithm proposed by this article and the test of applying it in ﬁnding minimal and maximal value. The third section introduces the sample data used by this article and the real case analysis. The fourth section proposes the research conclusion and suggestions. 2. A new Fruit Fly Optimization Algorithm 2.1. The basic concept of the Fruit Fly Optimization Algorithm The Fruit Fly Optimization Algorithm (FOA) is a new method for ﬁnding global optimization based on the food ﬁnding behavior of the fruit ﬂy. The fruit ﬂy itself is superior to other species in sensing
W.-T. Pan / Knowledge-Based Systems 26 (2012) 69–74
Fig. 1. Illustration of the body look of the fruit ﬂy and group iterative food searching of fruit ﬂy.
and perception, especially in osphresis and vision, which is as shown in Fig. 1. The osphresis organs of fruit ﬂies can ﬁnd all kinds of scents ﬂoating in the air; it can even smell food source from 40 km away. Then, after it gets close to the food location, it can also use its sensitive vision to ﬁnd food and the company’s ﬂocking location, and ﬂy towards that direction too. In this article, based on the food ﬁnding characteristics of the fruit ﬂy, it is divided into several necessary steps and a program example to be referred to by the readers. The steps are:
(2) Give the random direction and distance for the search of food using osphresis by an individual fruit ﬂy.
X i ¼ X axis þ RandomValue Y i ¼ Y axis þ RandomValue (3) Since the food location cannot be known, the distance to the origin is thus estimated ﬁrst (Dist), then the smell concentration judgment value (S) is calculated, and this value is the reciprocal of distance.
(1) Random initial fruit ﬂy swarm location is shown in the ﬁgure to the right of Fig. 1.
Dist i ¼
X 2i þ Y 2i
Si ¼ 1=Disti
InitX axis; InitY axis
-4.997 -4.998 -4.999 -5
Iteration Number Optimization process
Maximum Value 30
Iteration Number Fig. 2. Curve of extreme value solution of an iterative search solution.
W.-T. Pan / Knowledge-Based Systems 26 (2012) 69–74
Fig. 3. The search route chart for different numbers of fruit ﬂies.
2.2. Finding minimal value and maximal value example
[00031, 1]. After 100 times of iterative search of minimal value and maximal value, the program implementation result will gradually approach the solution of the functional extreme value. Fig. 2 is the curve drawn by the solution of the extreme value of iterative search function. From the above ﬁgure, it can be seen that the curve gradually approaches a functional minimal value of 00035, and the coordinate of the swarm of the fruit ﬂy is (76.9049, 82.8723); from the following ﬁgure, it can be seen that the curve gradually approaches a functional maximal value of 2, and the coordinate of the swarm of the fruit ﬂy is (80.4769, 75.2152). Further investigation will be performed in this section. In this article, how the number of fruit ﬂies affects the food searching efﬁciency will be investigated. From Fig. 3, it can be found that after 10 iterative searches, no matter in the search of minimal value or maximal value, the swarm with fewer fruit ﬂy numbers will have disadvantages of an unstable search route and a slower convergence speed; but the advantage is the execution speed will be faster. On the contrary, swarms with a large number of fruit ﬂies will have the advantage of a stable search route, and the convergence speed will be faster too; but the disadvantage is the program execution speed will be slower. Therefore, the researchers must consider the complexity of the optimization problem to select appropriately, the number of fruit ﬂies to perform the treatment for the problem.
In this article, FOA is used to get minimal value and maximal value with functions of respectively:
3. Case analysis
Y ¼ 00035 þ X 2
3.1. Sample data and variable
(4) Substitute smell concentration judgment value (S) into smell concentration judgment function (or called Fitness function) so as to ﬁnd the smell concentration (Smelli) of the individual location of the fruit ﬂy.
Smelli ¼ FunctionðSi Þ (5) Find out the fruit ﬂy with maximal smell concentration (ﬁnding the maximal value) among the fruit ﬂy swarm.
½bestSmellbestIndex0002 ¼ maxðSmellÞ (6) Keep the best smell concentration value and x, y coordinate, and at this moment, the fruit ﬂy swarm will use vision to ﬂy towards that location.
Smellbest ¼ bestSmell X axis ¼ XðbestIndexÞ Y axis ¼ YðbestIndexÞ (7) Enter iterative optimization to repeat the implementation of steps 2–5, then judge if the smell concentration is superior to the previous iterative smell concentration, if so, implement step 6.
and in the former one, the solution of minimal value is 00035, and in the latter, the solution of the maximal value is 2. The random initialization fruit ﬂy swarm location zone is [0, 10], the random ﬂy direction and distance zone of iterative fruit ﬂy food searching is
Financial distress historical data of 100 companies with stocks listed in the regular stock market and over-the-counter stock market from 2003 to 2004 are adopted as test data. Meanwhile, three ﬁnancial distress models such as fruit ﬂy algorithm optimized General Regression Neural Network (abbreviated as FOAGRNN),
W.-T. Pan / Knowledge-Based Systems 26 (2012) 69–74
Table 1 The descriptive statistical values of the ﬁnancial distress test data from 2003 to 2004. Factor
X1 X2 X3 X4
3798.39 296.67 57.79 197.05
000399.66 000399.78 00034344.32 00034641.7
53.455 3.3369 000367.8833 0003114.638
390.3753 46.02071 459.7565 643.7703
1645.88 918.9 55.9 2019.77
000399.41 000385.64 00033671.59 000323589.1
54.0975 6.9279 000356.4762 0003267.093
209.1663 95.54782 379.4832 2380.936
Fig. 4. The iterative RMSE trend of FOAGRNN search of optimization parameter; ROC curve of fruit ﬂy swarm location coordinate route and classiﬁcation prediction result.
General Regression Neural Network (abbreviated as GRNN) and Multiple Regression (abbreviated as MR) models are set up respectively to observe the efﬁcacy of fruit ﬂy algorithm optimization. The ﬁnancial distress historical data includes the growth force and proﬁtability force of the ﬁve ﬁnancial forces. The four ﬁnancial ratio index independent variables include revenue growth rate (X1), ﬁxed-asset growth rate (X2), operating proﬁt margin (X3) and proﬁt margin (X4), etc. The dependent variable Y is 50 companies in risk (represented by 1) and the paired 50 normal companies (represented by 0), which is used to construct and test the sample data of fruit ﬂy algorithm optimized General Regression Neural Network (abbreviated as FOAGRNN), General Regression Neural Network (abbreviated as GRNN) and Multiple Regression (abbreviated as MR). Table 1 shows the descriptive statistical values of four ﬁnancial ratio indexes.
3.2. The construction of three ﬁnancial distress classiﬁcation prediction models In this article, 100 sample data are normalized to make samples in the range from 0 to 1. It is then divided all sample data equally into ﬁve groups with each contains 20 data rows. Four groups of data used as training data to construct the model, and one group of data used as test data to test model stability. Thus, cross veriﬁcation is accomplished. The output result of three models is deﬁned as when it is smaller than or equal to 0.5, it is classiﬁed as 0 (that is, it is normal company); when it is larger than 0.5, it is classiﬁed as 1 (that is, company in risk). First, training data of four groups are used to construct three models of FOAGRNN, general GRNN and traditional Multiple Regression, respectively. For GRNN theories, the readers can refer to the related publications of
W.-T. Pan / Knowledge-Based Systems 26 (2012) 69–74
professor [23,24]. In the FOAGRNN model, this article has adopted the GRNN matlab toolbox and self-written MATLAB program. This article has also designed the spread parameter value of the FOA dynamic search of optimized GRNN. And ﬁnally, the FOAGRNN model is also constructed in this article. The initial value of the spread parameter of GRNN is set up in the range of [0.01, 1], and the neuron number of the network input layer is 4, and the output neuron number is one. In the initial parameter set up of the FOA, the random initialization fruit ﬂy swarm location range is [0, 100], the random ﬂy direction and distance zone of iterative fruit ﬂy food searching is [000310, 10], fruit ﬂy population size is 50, and iterative number is 1000. The way used in the FOA optimized GRNN is to ﬁrst calculate the distance between the individual location of the fruit ﬂies and the origin coordinate (0, 0), and then calculate the reciprocal. The smell concentration judgment value (S) is then calculated, and then substituted into the parameter spread of GRNN. After that, the training data is entered to get the network output value. Then with the target value, RMSE (or called Fitness) is calculated, the smaller the value, the better. Finally, the best smell concentration judgment value (S) is kept to be used as the spread value of GRNN, and the iterative search is made based on this method. Through the smell of the fruit ﬂy’s random food ﬁnding, and through the ﬂocking at the location of the highest concentration of smell using vision, the spread value of GRNN can be adjusted to its optimal value, and the RMSE between network output value and target value can be adjusted to the minimal value. The upper four ﬁgures in Fig. 4 represent the training data of 2003 and 2004, respectively. FOA is used to iteratively optimize the spread value of GRNN, the RMSE between generated output value and target value, and the trend of the fruit ﬂy swarm ﬂying route. After 1000 times of iterative evolution, in 2003, convergence can be seen in generation 183, with a coordinate of (3.5554, 3.5932), and the spread value and RMSE value is respectively [0.1978, 0.0177]; in 2004, convergence starts from generation 187 with a coordinate of (1.5845, 1.6680), spread value and RMSE value is respectively [0.4347, 8.5498e0003086]. However, the down two ﬁgures in Fig. 4 show that for the sample data from 2003 to 2004, four groups of the data are used as training data to construct the model, and one group of data is used as test data to cross-verify the results generated by three models such as FOAGRNN, general GRNN and traditional Multiple Regression model, and the results are drawn into an ROC curve. Bradley  pointed out that the larger the area above the reference line and underneath the curve, the better the classiﬁcation capability of the model. From the ﬁgure, it can be clearly seen that the FOAGRNN model has the best classiﬁcation capability. Then, from an observation of the ROC curve analysis output result through Table 2, we see that sensitivity (Sen) means the percentage of the number of the prediction result of 1 (that is, the company in risk) to the number of the actual value of 1. Speciﬁcity (Spe) means the percentage of the number of the prediction result of 0 (that is, the normal company) to the number of the actual value of 0; and professor  pointed out that Gini Index is 2 0004 AUC – 1. Here, these index values are the larger the better. As shown in the table, for 2003 FOAGRNN model, the Sen is 0.940, Spe is 0.920, area un-
Table 2 ROC curve analytic index value of the test results of three models.
FOAGRNN GRNN MR FOAGRNN GRNN MR
0.940 0.840 0.820 0.920 0.800 0.800
0.920 0.820 0.720 0.940 0.800 0.760
0.930 0.830 0.770 0.930 0.800 0.780
0.860 0.660 0.540 0.860 0.600 0.560
der the curve (AUC) is 0.930, and Gini Index is 0.860; for 2004 FOAGRNN model, the Sen is 0.920, Spe is 0.940, the area under curve (AUC) is 0.930, and the Gini Index is 0.860, which are all higher than those of the general GRNN and traditional Multiple Regression model. Therefore, the FOAGRNN model has a very good classiﬁcation prediction capability. 4. Conclusion The major contribution of this article is to propose a new Fruit Fly Optimization Algorithm (FOA) with a real application in ﬁnding maximal value and minimal value. From the test result, it is found that the FOA can usually ﬁnd solutions correctly, and the stability of the fruit ﬂy swarm search route is obviously related to fruit ﬂy quantity. In addition, this article further adopts ﬁnancial distress historical data in the Taiwan area to test the optimization capability of the FOA. From the analysis result, it can be seen that through the FOA, the spread value of the GRNN network parameter can be optimized, and the classiﬁcation prediction capability of the GRNN can obviously be enhanced. This algorithm not only has the feature of being easy to understand, but also is easy to be written into program code. Meanwhile, the program code, as compared to other algorithms, is not too long, and it is thus easy to be used to handle all kinds of optimization problems. However, the brand new algorithm as proposed by the FOA for this article might not be thorough, and we thus hope that more researchers can participate in the promotion and test. References  D.R. Anderson, C.L. Moodie, Optimal buffer storage capacity in production line, Int. J. Prod. Res. 7 (1969) 233–240.  Y.D. Bertrand, D. Barba, Feature selection by a genetic algorithm application to seed discrimination by artiﬁcial vision, J. Science Food Agric. 76 (1998) 77–86.  A.P. Bradley, The use of the area under the ROC curve in the evaluation of machine learning algorithms, Pattern Recognit. 30 (7) (1997) 1145–1159.  B. Bullnheimer, R. Hartl, C. Strauss, An improved ant system algorithm for the vehicle routing problem, Ann. Oper. Res. 89 (1999) 319–328.  V. Cho, MISMIS – A comprehensive decision support system for stock market investment, Knowl. Based Syst. 23 (6) (2010) 626–633.  M. Dorigo, L.M. Gambardella, Ant colony system: a cooperative learning approach to the travelling salesman problem, IEEE Trans. Evol. Comput. 1 (1) (1997) 53–66.  M. Friedman, A mathematical programming model for optimal scheduling for buses departure under deterministic condition, Transp. Res. 10 (2) (1976) 83– 90.  Y. Fukuyama, H. Yoshida, 2001. A particle swarm optimization for reactive power and voltage control in electric power systems. Congress Evolutionary Computation 1, 87–93.  S.B. Gershwin, An efﬁcient decomposition method for the approximate evaluation of tandem queues with ﬁnite storage space and blocking, Oper. Res. 35 (1987) 291–305.  E. Hadavandi, H. Shavandi, A. Ghanbari, Integration of genetic fuzzy systems and artiﬁcial neural networks for stock price forecasting, Knowl. Based Syst. 23 (8) (2010) 800–808.  D.J. Hand, R.J. Till, A simple generalisation of the area under the ROC curve to multiple class classiﬁcation problems, Mach. Learn. 45 (2) (2001) 171–186.  P.E. Hart, N.J. Nilsson, B. Raphael, A formal basis for the heuristic determination of minimum cost paths, IEEE Trans. Syst. Sci. Cybern. 4 (2) (1968) 100–107.  N. Koncz, J. Greenﬁeld, K. Mouskos, A strategy for solving static multipleoptimal-path transit network problems, J. Transp. Eng. 122 (3) (1996) 218– 225.  Y.W. Leung, Y. Wang, An orthogonal genetic algorithm with quantization for global numerical optimization, IEEE Trans. Evolution. Comput. 5 (2001) 41–53.  H. Li, J. Sun, Ranking-order case-based reasoning for ﬁnancial distress prediction, Knowl.Based Syst. 21 (8) (2008) 868–878.  W.T. Pan, 2011. A new evolutionary computation approach: Fruit Fly Optimization Algorithm, 2011 Conference of Digital Technology and Innovation Management, Taipei. Program code on the website
W.-T. Pan / Knowledge-Based Systems 26 (2012) 69–74
 Roman M. Balabin, Ravilya Z. Saﬁeva, Biodiesel classiﬁcation by base stock type (vegetable oil) using near infrared spectroscopy data, Anal. Chim. Acta 689 (2) (2011) 190–197.  Roman M. Balabin, Ekaterina I. Lomakina, Support vector machine regression (SVR/LS-SVM)—an alternative to neural networks (ANN) for analytical chemistry? Comparison of nonlinear methods on near infrared (NIR) spectroscopy data, Analyst 136 (2011) 1703–1712.  Roman M. Balabin, Ravilya Z. Saﬁeva, Ekaterina I. Lomakina, Near-infrared (NIR) spectroscopy for motor oil classiﬁcation: from discriminant analysis to support vector machines, Microchem. J. 98 (1) (2011) 121–128.
 F.J.M. Salzborn, Optimum bus scheduling, Transp. Sci. 6 (2) (1972) 137–148.  D.F. Specht, Probabilistic neural networks and the polynomial adaline as complementary techniques for classiﬁcation, IEEE Trans. Neural Netw. 1 (1) (1990) 111–121.  D.F. Specht, A general regression neural network, IEEE Trans. Neural Netw. 2 (6) (1991) 568–576.  D. Srinivasan, W.H. Loo, R.L. Cheu, 2003. Trafﬁc incident detection using particle swarm optimization. Swarm Intelligence Symposium, Proceedings of the 2003 IEEE, pp.144–151.
Taking The Fifth Pdf Free Download Windows 10
Patient-centered, high-quality health care relies on the well-being, health, and safety of health care clinicians. However, alarmingly high rates of clinician burnout in the United States are detrimental to the quality of care being provided, harmful to individuals in the workforce, and costly. It is important to take a systemic approach to address burnout that focuses on the structure, organization, and culture of health care.
Taking The Fifth PDF Free Download
Taking Action Against Clinician Burnout: A Systems Approach to Professional Well-Being builds upon two groundbreaking reports from the past twenty years, To Err Is Human: Building a Safer Health System and Crossing the Quality Chasm: A New Health System for the 21st Century, which both called attention to the issues around patient safety and quality of care. This report explores the extent, consequences, and contributing factors of clinician burnout and provides a framework for a systems approach to clinician burnout and professional well-being, a research agenda to advance clinician well-being, and recommendations for the field.