Object classification(Framework & any two techniques)
The purpose of object classification is to provide necessary information on the object classes that are 
used as a uniform classification for identifying the transactions of the Federal Government by the nature of the goods or services purchased. 
Every obligation recorded by the Department of 
Defense must be coded with an object class.
These obligations by object class must be accumulated and reported to the Treasury on a quarterly basis. 

a. Obligations incurred are the amounts of 
orders placed, contracts awarded, services received, and similar transactions during an accounting period that will require payment during the same or a future period. Such amounts include payments for which obligations have not been 
previously recorded, along with adjustments for differences between obligations previously recorded and actual payments to liquidate those obligations. The installation-level budgetary 
account structure requires that the amount of oblig
ations incurred be segregated into undelivered 
orders and accrued expenditures - unpaid or paid. 

b. Obligations are classified by the initial purpose for which they are incurred, rather than for the end product or service provided. For example,if payments are made directly for personal services or supplies used in the construction 
of a building, the amounts should be classified as 
obligations for personnel compensation or supplies rather than for land and structures. Conversely, if contracts are awarded for the construction of a building, the amounts should be classified in the object class for lands and structures rather than inthe object class used to report individual types of 
materials and services required to construct the building. 

Techniques
1. K Nearest Neighbour

Nearest neighboursalgorithm is considered as statistical learning algorithms and it is extremely simple to implement and leaves itselfopen to a wide variety of variations. 
In brief, the training portion of nearest-neighbourdoes little more than store the data points presented toit. When asked to make a prediction about an unknown point, the nearest neighbour classifier finds the closest training-point to the unknown point and predicts the category of that training point accordingly to some distance metric. 

The distance metric used in nearest neighbour methods for numerical attributes can be simple Euclidean distance. In k nearest neighbour algorithm 
to memorize the all data and then find the closest point. The neighbour vote for the label is vote (+) = 2 and (-) = 1. 

2. Adaboost Classifiers 
Adaboost (Adaptive boosting) is a machine learning algorithm. It can be used with many different classifiers to improve the accuracy. Adaboost is adaptive in the sense that subsequent weak learners are tweaked. 
Adaboost focuses on more previously misclassified samples.Initially all samples are equal weights. Weight may change at each boosting round. 
It can be less susceptible to the over fitting problem than other learning algorithms. The individual learners can be weak, but as long as the performance of each one is slightly better and the final model can be proven to converge to a strong learner. Steps of adaboost classifiers are Bootstrapping, Bagging, Boosting, and Adaboost.