This paper explains the problem of determinizing probabilistic data which takes only un deterministic input. While accepting those types of un deterministic input data the legacy systems stores those type of data. The automated data analysis/enrichment techniques generate probabilistic data. These techniques are entity resolution, information extraction, and speech processing. The legacy application system may correspond to the existing web applications like Flicker, Picasa, etc. The goal of this survey is to generate a deterministic representation of probabilistic data which optimizes the deterministic data quality on the end-application. Those determinization problems should be deployed in the area of two different data processing tasks they are triggers and selection queries. The top-1 selection techniques are traditionally used for determinization problem which create suboptimal performance for many applications. Instead this survey proposes a query-aware strategy and explains its advantages over existing solutions through a comprehensive empirical study over real and synthetic datasets. We propose a probabilistic topic model for analyzing and extracting content-related annotations from noisy annotated discrete data such as web pages stored in social bookmarking services. In these services, since users can attach annotations freely, some annotations do not describe the semantics of the content, thus they are noisy, i.e. not content-related. The extraction of content-related annotations can be used as a preprocessing step in machine learning tasks such as text classification and image recognition, or can improve information retrieval performance. The proposed model is a generative model for content and annotations, in which the annotations are assumed to originate either from topics that generated the content or from a general distribution unrelated to the content. We demonstrate the effectiveness of the proposed method by using synthetic data and real social annotation data for text and images.