B. Chandra

Moving towards efficient decision tree construction

Motivated by the desire to construct compact (in terms of expected length to be traversed to reach a decision) decision trees, we propose a new node splitting measure for decision tree construction. We show that the proposed measure is convex and cumulative and utilize this in the construction of decision trees for classification. Results obtained from several datasets from the UCI repository show that the proposed measure results in decision trees that are more compact with classification accuracy that is comparable to that obtained using popular node splitting measures such as Gain Ratio and the Gini Index.

Fuzzifying Gini Index based decision trees

Crisp decision tree algorithms face the problem of having sharp decision boundaries which may not be found in all real life classification problems. A fuzzy decision tree algorithm Gini Index based (G-FDT) is proposed in this paper to fuzzify the decision boundary without converting the numeric attributes into fuzzy linguistic terms. Gini Index is used as split measure for choosing the most appropriate splitting attribute at each node. The performance of G-FDT algorithm is compared with Gini Index based crisp decision tree algorithm (SLIQ) using several real life datasets taken from the UCI machine learning repository. G-FDT algorithm outperforms its crisp counterpart in terms of classification accuracy. The size of the G-FDT is significantly less compared to SLIQ.

An open and safe nested transaction model: concurrency and recovery

In this paper, we present an open and safe nested transaction model. We discuss the concurrency control and recovery algorithms for our model. Our nested transaction model uses the notion of a recovery point subtransaction in the nested transaction tree. It incorporates a prewrite operation before each write operation to increase the potential concurrency. Our transaction model is termed ‘‘open and safe’’ as prewrites allow early reads (before writes are performed on disk) without cascading aborts. The systems restart and buAer management operations are also modeled as nested transactions to exploit possible concurrency during restart. The concurrency control algorithm proposed for database operations is also used to control concurrent recovery operations. We have given a snapshot of complete transaction processing, data structures involved and, building the restart state in case of crash recovery. ” 2000 Elsevier Science Inc. All rights reserved.

Crash Recovery in an Open and Safe Nested Transaction Model

In this paper, we present an open and safe nested transaction model and discuss the crash recovery issues. We introduce the notion of a recovery point subtransaction in a nested transaction tree. We introduce prewrite operations to increase concurrency. Our model is open and safe as prewrites allow early reads (before database writes on disk) without cascading aborts. The systems restart and buffer management operations are modeled as nested transactions to exploit possible concurrency during restart. Our model is useful in handling long-duration transactions.

Virtual partition algorithm in a nested transaction environment and its correctness

Abstract In this paper, we present a formal description of the virtual partition algorithm in a nested transaction environment and prove its correctness. We model the virtual partition algorithm in a nested transaction environment using the I/O automaton model. The formal description is used to construct a complete correctness proof that is based on standard assertional techniques and on a natural correctness condition, and takes advantage of the modularity that arises from describing the algorithm as nested transactions. Our presentation and proof treat issues of data replication entirely separately from issues of concurrency control. Moreover, we have identified that the virtual partition algorithm cannot be proven correct in the sense of Goldmans work [ACM Trans. Database Syst. 19(4) (1994) 537] on Giffords quorum consensus algorithm using the serializability theorem defined by Fekete et al. [Atomic Transactions, Morgan-Kaufmann, USA, 1994]. Thus, we have stated a weaker notion of correctness conditions, which we call the reorder serializability theorem. We have shown that not all classes of replication algorithms can be proven in the way Goldman has presented the proof of Giffords quorum consensus algorithm.

On the Correctnes of Virtual Partition Algorithm in a Nested Transaction Environment

In this paper, we model the virtual partition algorithm in a nested transaction environment using I/O automaton model. The formal description is used to construct a complete correctness proof that is based on standard assertional techniques and on a natural correctness condition, and takes advantage of modularity that arises from describing the algorithm as nested transactions. Our presentation and proof treat issues of data replication entirely separately from issues of concurrency control. Moreover, we have identified that virtual partition algorithm can not be proven correct in the sense of Goldmans work [7] on Giffords Quorum Consensus Algorithm using the serializability theorem defined by Fekete et al.[4]. Thus, we have stated a weaker notion of correctness conditions, which we call reorder serializability theorem.