This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
A balance scale is an tool that measures the weight or the mass of the object. Balance the touchstone weight to an unknown weight using a horizontal lever is called balance. Balance scale are measured in many products like industrial and commercial applications from feahers to loaded tractors. For measuring the body weight of human beings a special medical scales and bathroom scales are used.
The first measuring tool which was fantasy was the balance scale. The balance scale consist of the two alike lengths of the horizontal lever arms which is also called the weighing pan with a beam this is also called as the scale pan. This pan is suspended from each other. The mass which we have to measure is to be found in one of the pan and the pattern masses by which we are measuring the mass is placed in the other pan this is to done till the beam comes close to the equilibrium as much as potential A slider weight is move with a modify scale in the precision balances. really the balance scale is technically match up to with the weights not with the masses. The mass is proportional to the weight of the object. The measured weights are used with the balances are frequently characterized in the mass units.
For precision mass standards the balances are used because the spring scales which are unlike is different in the local gravity because the accuracy of the spring scales is not affected which is almost vary at the different places on the earth by the percentage of almost 0.5. By moving the balances to another place there will be a change in the strength of the gravitational field. These balances will not change the standard mass because the balances beam is affected equally in the moments of force of both the sides.
The balances hinge is fundamentally friction free in very a small amount of measurements which are achieve by ensure the balances. It amplify the deviations from the location of the balance by attach the pole to the balance beam finally it allows the actions of a small mass to be useful by use the lever principle which allow fractional masses. And also it calculated with the measuring arm of the beam which is the utmost accuracy describe above the densities of the masses are involve in air whose effects are charge on the need to be an grant for the enthusiasm.
The centre of the fulcrum in the balance is consisting of the original form of the balance beam. The sharp v-shaped placed in a shallower v-shaped bearing can be consist of a fulcrum for getting the highest accuracy. The combination of the different masses are hanged on the one of the end of the beam to get the mass of the object while the object of the mass of which has to measured was hanged on the other side of the balance scale. The center beam balance is the most accurate technology available for the high precision work and this technology which now a days we are using for calibrating the test weights is very common.
The off centre beam is one of the large reference mass which we can use to reduce the need. With the off centre beam the balance which we measure will be almost same as the scale with a centre beam for getting the accuracy of the contents which are simply swapped the off centre beam requires the special reference. The poise which is also called as the sliding weight can be installed in the calibration procedure for reducing the need for the small graduate reference but the one thing we should keep in the mind that accurate lever ratio of the mass beam should be adjusted in the accurate mass of the poise.
Sources of errors :-
An old two pan balances.
Below causes the error in the source of the potential which is due to the high precision.
- Buoyancy is an error we get due to the object we are weight leaks some air, we can balance in the vacuum some of the high precision.
- We will found some error in reference weight.
- Air gusts even small ones which push the scale up or down.
- There is also one of the small errors called as Air Gusts by which the balance scale will go up and down.
- In the moving components the friction can prevent the error in the scale from approaching equilibrium.
- To the weight which we measure the settling airborne can be dust contribution.
- Due to the miss calibration overtime the temperature or the accuracy will change.
- Due to the thermal expansion contraction of components the mechanical components are mis aligned.
- On the ferrous components the magnetic fields are acting.
- From the feet shuffled on the carpets and on a dry day the electro static fields are forces.
- Between the air and the substances the chemical reactivity has been weighed.
- There will be a condensation on the cold items because of the atmospheric water.
- From the wet items there will be a evaporation of water.
- From hot or cold items in the balance scale there will be a convection of air.
- There will be a disturbance like vibration in the scale and seismic with the scale because if there be any passing vehicles or rumbling from the balance scale for example trucks.
The Weka software consists of number of collection example visualization tools and algorithms which interface bb 11 programming language. The original non java version of the java was designed to final the data weak software is a freely domains it is freely vial able for the public because it doesn't ask for the licence and vey portable it weka software runs on java programming language work on old or new system it id very easy to use the modelling techniques and the data possessing
The very good support by the weka software is the data processing techniques predicted by the weka software are depends on assumptions weka software is not applicable for milt relation and suitable for the singe data base by using java data base connectivity the weka software provide way to SQL language the fixed attributes are described by the each number of the data points
The same functionality can be accessed by the components of the based knowledge of the flow interface but essentially the weka's main user interface is the explorer. And also it can be accessed from the command line. The weka's machine learning algorithms on a collection of the number of the data base is also the experimenter
The weka explorer interface has number of panels to access the main components of the work bench the data base has important data which has the good facilities for preprocess panel, filtering file is used for csv files and for the preprocessing of the data the filter can be used to delete the instances and attributes according to the specific steps filters are also used for transferring the data. The filters also help with the classification and regression algorithms to visualize errors. Classify panels will give access to rule learners who try to think the impotence between data and attributes.
- The weka software was developed in 1993 by waikato university which was the original version in New Zealand and it was redeveloped in 1997 decision taken to scratch in java and for better modeling algorithms.
- The weka software was also awarded by SIGKDD(data mining and the knowledge discovery service) in 2005 and pentaho corporation award in 2006.
- The weka software is one of the top most ranking web side which is sourcefog.net until 11-06-2009
For create and just beginning the choice tree the data mining researchers have residential the amount of algorithms which will be by means of with the help out of weka that is j4.8 decision learner. This is zero but the C4.5 which is as well called as WEKA'S implementation of the landmark choice tree program. C4.5 was developed 20 years of period reverse opening in the late 1970's by J. Ross Quilan. . After the consideration it turn out that the C4.5 and J4.8 decision tree beginner can also be conventional in arithmetic what and the attributes as well as the nominal attributes, so that we can use our original FishersIrisDataset.arff file, which also has mathematical values for the flower width and its attributes.
The common rescaling which is used to put in weka classifiers and trees with the help of J4.8 intend to the use of proper model size in the regular and apposite to the binomial and making the amount up to 0.5 continuity of modification and consistent with the data set. And the minimum base leaf count option on UN weight number of count.
On top of these are the simple changes which are making prune more competent and effective prejudiced data, and also it helps to decrease apparent over right. This should have to be the coop where either the weights must reflect the lost principles charge or survey-sampling probabilities. The alteration of the algorithms J4.8 states that it would not have work on its own. In contrast the algorithm J4.8. The allocation for the this algorithm had been written to maintain single set of counts only and also to intolerant data statically algorithms which is often have need of for both weighted and unwished number of count.
Below is the screen shot of the weka software when the Balance Scale data set is uploaded.
This very method. The data is very much different from other data set and the data showed in the fishers' article. The 35th of the sample can be 4.9,3.1,1.5 and 0.2 balance scale in where the error is the fourth feature. And the 38th sample is 4.9,3.6 and 1.4 where the errors can also be in the second and the third features also.
The following is the Attribute information of the balance scale.
- Sepal length in cm
- Sepal width in cm
- Petal length in cm
- Petal width in cm
Below is the screen shot which shows the total number of instance, confusion matrix and also the percentage when J48 has been applied to balance scale.
A puzzlement matrix contains in sequence about real and predicated classifications complete by a classification system. The confusion Matrix show that there are 15 scale in the test set in the middle of all of which were very much correctly classified as balance. There were also nineteen Vesicular in the test set, which were all confidential correctly. There were also seventeen Virginia, but two were classified not accurately as it was vermicular by our decision tree.
The following is the decision tree I get after Appling the Iris information set to the WEKA software and ran J4.8 algorithms on it.
a b c <-- classified as
38 9 2 | a = B
10 274 4 | b = R
20 4 264 | c = L
l a is the correct classification as first class
l b is the number of incorrect classification of the first class
l c is the number of incorrect classification of the second class
l d is the correct classification at the second class
The Decision Tree is like a supporter tool that uses a tree-like graph of all the statement and also their achievable punishment. These consist of possibility knowledge conclusion and also the property expenditure and their efficiency. The Decision trees are also regularly used in function research where wholly in close stop operational so that it can assist to recognize a plan of most probable to reach a objective single more practice of choice tree is that a evocative means for the manipulative of provisional probability.
The neural network was started to present he how the mind of the human will performed. With the help of the perception and the machine called b-type some of the ideas has been started to apply on the computational models. The neural network was stared in the year 1809
In the year 1975 cognitron was the first multilayered which was developed by the neural network the methods which are used to set the inter connection of the weights of the main structure of he neural network changes from one neural network to another neural network. neural network can gather the information in the one direction not more than one but it is possible when they can go back to its same place and to the fourth place until neural go on the final state and the self activation at a node occurs.
In 1975 cognitron 1st layered developed by neural network and connection of the weights structure methods .the later discovery of biological models in to the fields.
Neural network can gather one direction not more than one but it is possible when they can go on the final state
Applications of Neural Networks:-
To infer a function from observation and also to start it the utility of the neural network models lies in the fact in which they can be. Where the complexity of the data or tasks makes the design neural network is particularly useful for the functions by hand impractical.
The neural network includes some of the system identification in some of the application i,e the controlling of the vehicles, controlling of the process. Playing the games and making the decision pattern recognition and the last at the least the sequence recognition medical diagnosis it also control the financial application, the data mining visualization and the filtering of the spam which we will see in our e-mails.
To simulate the research and to apply the artificial neural network in biological neural network. The neural network software is used. And in some cases a wider array of adaptive system is used in the place of neural network. There are three types of major learning paradigms each corresponding to a particular abstract learning which are as follows.
- Supervised learning.
- Unsupervised learning and.
- Reinforcement learning.
Usually any given type of network architecture can be employed in any of those tasks.
Below is the Neural Networks diagram for the Balance Scale which I got after applying the data in to the weka.
- The correctness of balance scale using j48 algorithm is 90% with an instance of 139'
- the correctness of instances when applied on neural network which is also called as multilayered perceptron is 93% which is nearly same as the j48.
- The total number of instances using both j48 and neural network is 625
- I came to the conclusion from results of both j48 and neural network that we can use any of the algorithms to perform this operation.