• No results found

Machine Learning Based Predictive Analysis of Tariff Rate Prediction

N/A
N/A
Protected

Academic year: 2022

Share "Machine Learning Based Predictive Analysis of Tariff Rate Prediction"

Copied!
5
0
0

Loading.... (view fulltext now)

Full text

(1)

Volume 6, Issue 03, March 2020 (ISSN: 2394 – 6598)

181

©IJETIE 2020

Machine Learning Based Predictive Analysis of Tariff Rate Prediction

Arul Kumar V1, Agalya S2, Gokul Chakkaravarthi S2, Gokula Priya E2

1 Assistant Professor, Department of CSE, Sri Krishna College of Engineering and Technology, Coimbatore.

2 Department of CSE, Sri Krishna College of Engineering and Technology, Coimbatore.

ABSTRACT

A tariff rate is a price at which a certain cargo is delivered from one point to another. The price depends on the form of the cargo, the mode of transport, the weight of the cargo, and the distance to the delivery destination.

Simple mean applied tariff is the unweight average of effectively applied rates for all products subject to tariffs calculated for all traded goods. Less-than-truckload (LTL) is used for smaller truck freight loads. In single trucks LTL carriers hold several shipments for numerous customers. Combining several cargo owner’s cargoes in one truck reduces the cost each cargo owner must pay. Shipping by LTL offers significant savings over shipping the same load in a dedicated truck. The details were listed at six- or eight-digit point utilizing the Harmonized Trading Framework. Tariff line results have been combined with Standard International Trade Classification (SITC) revision 3 codes to identify product groupings. Specific rates have been converted to their ad valorem equivalent rates and have been included in the calculation of simple mean tariffs.to take the truck traveling details and it traveling cost for predict that truck overall traveling cost with tariff rate. We can collect the details from real time field. Applying regression method for prediction, here we apply linear regression, support vector regression, random forest regression, and decision tree regression. Using this all algorithm to find best one for prediction.

KEYWORDS: Tariff, Classification, Prediction, Algorithm, Regression, Machine Learning.

1. INTRODUCTION

In system mastering, a hyper parameter is a configuration variable that’s outside to the model and whose price is not envisioned from the records given. Hyper parameters are a critical part of the process of estimating version parameters and are often described via the practitioner. When a system mastering set of rules is used for a particular trouble, along with the usage of a grid seek or a random seek set of rules, then you’re actually tuning the hyper parameters of the model to discover the values that bring about the most accurate predictions.

Implementing a tariff is the most common way of

defending one's economy from import competition.

In general terms, a tariff is any tax or charge that a government receives. Alternatively, as in the railroad tariffs, the word "tariff" is used in a nontrade context. The word, however, is used much more frequently to refer to imported goods tax. Tariffs have been applied by countries for decades and were one of the most widely used tools for government tax collection. It is primarily because it is fairly easy to position customs officials at a country's border and collect a fee on goods that enter.

Administratively, one of the simplest taxes to raise possibly is a tariff. High tariffs can, of course, induce the smuggling of goods through non-

(2)

182

©IJETIE 2020 traditional entry points, but here we will ignore that

issue.In an international trade course, tariffs are worth identifying early because adjustments in tariffs are the primary way in which countries can liberalize trade or defend their economies. However, this is not the only way, because countries are still introducing tariffs, quotas and other forms of regulations that can impact trade flows between governments.All such approaches are later to be described and addressed, but for now it is enough to understand tariffs as they still reflect the basic policy that affects international trade patterns.As people speak about trade liberalization, they typically mean lowering tariffs on imported goods, enabling the products to enter at lower prices. Since reducing the cost of trade makes it more competitive, trade should become freer. What economists and others mean by free trade is a complete removal of tariffs and other barriers to trade. Any increase in tariffs, on the other hand, is referred to as protectionism. Since tariffs boost the cost of importing goods from abroad but not from domestic firms, they shield domestic firms competing with imported products. Such domestic companies are called suppliers of the imports. Less than truck cargo transportation, or more generally known as LTL, is used for small freight transport.

This is one of the prime options when freight does not require a full trailer to be used. Businesses use the LTL service for smaller shipments that only require less than full truckload. As a common procedure, LTL combines shipping with other small vessels to fill a truck. When shipping LTL, you just pay for a portion of a regular truck trailer which is filled by their freight. The unoccupied room is then filled by other shippers and their shipments. This form of shipping is most effective when the freight weighs between 150 and 15,000 pounds. Generally, LTL shipping is almost acceptable to all forms of cargo-service businesses. But because of varying requirements that different people need, LTL

shipping would be best for businesses that have freight below 15,000 pounds and does not require a fully loaded trailer shipping is also the optimal choice for peoples demanding maximized cost savings shipping is the best option. For this, a shipment can be shipped as easily as possible.

1. MACHINE LEARNING

Machine Learning is a subset of Artificial Intelligence (AI) and is based on the idea that machines should be given the access to data, and should be left to learn and explore for themselves.

It deals with the extraction of patterns from large data sets. Machine learning enables a machine to automatically learn from data, improve performance from experiences, and predict things without being explicitly programmed

a. LINEAR REGRESSION:

Linear regression attempts to model the relationship between two variables by fitting a linear equation to observed data. One variable is considered to be an explanatory variable, and the other is considered to be a dependent variable. For example, a modeler might want to relate the weights of individuals to their heights using a linear regression model. Before attempting to fit a linear model to observed data, a modeler should first determine whether or not there is a relationship between the variables of interest. This does not necessarily imply that one variable causes the other (for example, higher SAT scores do not cause higher college grades), but that there is some significant association between the two variables. A scatterplot can be a helpful tool in determining the strength of the relationship between two variables. If there appears to be no association between the proposed explanatory and dependent variables (i.e., the scatterplot does not indicate any increasing or decreasing trends), then fitting a linear

(3)

183

©IJETIE 2020 regression model to the data probably will not

provide a useful model. A valuable numerical measure of association between two variables is the correlation coefficient, which is a value between -1 and 1 indicating the strength of the association of the observed data for the two variables. A linear regression line has an equation of the form Y = a + bX, where X is the explanatory variable and Y is the dependent variable. The slope of the line is b, and a is the intercept (the value of y when x = 0).

b. REGRESSION TREES:

Decision tree builds regression or classification models in the form of a tree structure.

It breaks down a dataset into smaller and smaller subsets while at the same time an associated decision tree is incrementally developed. The final result is a tree with decision nodes and leaf nodes. A decision node (e.g., Outlook) has two or more branches (e.g., Sunny, Overcast and Rainy), each representing values for the attribute tested. Leaf node (e.g., Hours Played) represents a decision on the numerical target. The topmost decision node in a tree which corresponds to the best predictor called root node. Decision trees can handle both categorical and numerical data. The core algorithm for building decision trees called ID3 by J. R.

Quinlan which employs a top-down, greedy search through the space of possible branches with no backtracking. The ID3 algorithm can be used to construct a decision tree for regression by replacing

Information Gain with Standard

Deviation Reduction.

c. RANDOM FOREST REGRESSION:

The Random Forest is one of the most effective machine learning models for predictive analytics, making it an industrial workhorse for machine learning. The random forest model is a type

of additive model that makes predictions by combining decisions from a sequence of base models. More formally we can write this class of models as: g(x)=f0(x)+f1(x)+f2(x)+... Different kinds of models have different advantages. The random forest model is very good at handling tabular data with numerical features, or categorical features with fewer than hundreds of categories.

Unlike linear models, random forests are able to capture non-linear interaction between the features and the target. One important note is that tree based models are not designed to work with very sparse features. When dealing with sparse input data (e.g.

categorical features with large dimension), we can either pre-process the sparse features to generate numerical statistics, or switch to a linear model, which is better suited for such scenarios.

D. SUPPORT VECTOR REGRESSION:

Support Vector Machines are very specific class of algorithms, characterized by usage of kernels, absence of local minima, sparseness of the solution and capacity control obtained by acting on the margin, or on number of support vectors, etc.

They were invented by Vladimir Vapnik and his co- workers, and first introduced at the Computational Learning Theory (COLT) 1992 conference with the paper. All these nice features however were already present in machine learning since 1960’s: large margin hyper planes usage of kernels, geometrical interpretation of kernels as inner products in a feature space. Similar optimization techniques were used in pattern recognition and sparsness techniques were widely discussed. Usage of slack variables to overcome noise in the data and non - separability was also introduced in 1960s. However it was not until 1992 that all these features were put together to form the maximal margin classifier, the basic Support Vector Machine, and not until 1995 that the soft margin version was introduced. Support

(4)

184

©IJETIE 2020 Vector Machine can be applied not only to

classification problems but also to the case of regression. Still it contains all the main features that characterize maximum margin algorithm: a non- linear function is leaned by linear learning machine mapping into high dimensional kernel induced feature space. The capacity of the system is controlled by parameters that do not depend on the dimensionality of feature space. In the same way as with classification approach there is motivation to seek and optimize the generalization bounds given for regression. They relied on defining the loss function that ignores errors, which are situated within the certain distance of the true value. This type of function is often called – epsilon intensive – loss function. The figure below shows an example of one-dimensional linear regression function with – epsilon intensive – band. The variables measure the cost of the errors on the training points. These are zero for all points that are inside the band.

2. PROPOSED FRAMEWORK

First step to split data into train and test for finding an accuracy for our data and model this step performed by train and test split module file. Its taken from sklearn packages. Data Pre- processing for Machine learning in Python. Pre- processing refers to the transformations applied to our data before feeding it to the algorithm. Data Pre- processing is a technique that is used to convert the raw data into a clean data set. In python, scikit-learn library has a pre-built functionality under sklearn.

pre-processing. There are many more options for pre-processing which we’ll explore. To convert a string object to binary values that means convert bytes. Then take the input for binary values to process and convert decimal then it will consider for applying algorithm Here we take the input of Main hub and local hub values for feature extraction.

Analysing the columns and comparing equalling variables or columns Using linear regression to

predict and print the test data output for our reference Then comparison chart will be shown output.

3. Dataset description:

There are using 14 columns for analysing and prediction to take a class or label in last column total remaining all are features. From, to and location this feature are converted into decimal values for process using distances to find fuel littler then find fuel cost based on distance and toll gate to find tariff of that truck. Using all the features value to find the label columns means find total amount

Fig.1. Proposed Framework

Fig.2. Distance Measurement

(5)

185

©IJETIE 2020 Fig.3. Rate prediction count vs cost

Fig.4. Tariff Prediction

Fig.5. Tollgate Prediction 4. Conclusion

Specific rates have been converted to their ad valorem equivalent rates and have been included in the calculation of simple mean tariffs.to take the truck traveling details and it traveling cost for predict that truck overall traveling cost with tariff

rate. We can collect the details from real time field.

Applying regression method for prediction, here we apply linear regression, support vector regression, random forest regression, and decision tree regression. Using this all algorithm to find best one for prediction.

REFERENCES

[1] Corinna Cortes and Vladimir Vapnik. Support- vector networks. Machine Learning, 20, 1995.

[2] Yoav Freund and Robert E. Schapire. A decision-theoretic generalization of on-line learning and an application to boosting. In Computational Learning Theory: Eurocolt ’95, pages 23–37.

Springer-Verlag, 1995.

[3] G. John, R. Kohavi, and K. Pfleger. Irrelevant features and the subset selection problem. In Machine Learning Conference, pages 121–129.

Morgan Kaufmann, 1994.

[4] H. Rowley, S. Baluja, and T. Kanade. Neural network-based face detection. In IEEE Patt. Anal.

Mach. Intell., volume 20, pages 22–38, 1998.

[5] R. E. Schapire, Y. Freund, P. Bartlett, and W. S.

Lee. Boosting the margin: a new explanation for the effectiveness of voting methods. Ann. Stat., 26(5):1651–1686, 1998.

[6] Robert E. Schapire and Yoram Singer. Improved boosting algorithms using confidence-rated predictions. Machine Learning, 37:297–336, 1999.

[7] H. Schneiderman and T. Kanade. A statistical method for 3D object detection applied to faces and cars. In Computer Vision and Pattern Recognition, 2000.

[8] Paul Viola and Michael J. Jones. Robust real- time object detection. In Proc. of IEEE Workshop on Statistical and Computational Theories of Vision, 2001.

References

Related documents