Ethnic differences in grains consumption and their contribution to intake of B-vitamins: results of the Multiethnic Cohort Study

Background: Research indicates that a diet rich in whole grains may reduce the risk of prevalent chronic diseases, including cardiovascular disease, diabetes, and some cancers, and that risk for these diseases varies by ethnicity. The objective of the current study was to identify major dietary sources of grains and describe their contribution to B vitamins in five ethnic groups. Methods. A cross-sectional mail survey was used to collect data from participants in the Multiethnic Cohort Study in Hawaii and Los Angeles County, United States, from 1993 to 1996. Dietary intake data collected using a quantitative food frequency questionnaire was available for 186,916 participants representing five ethnic groups (African American, Latino, Japanese American, Native Hawaiian and Caucasian) aged 45-75 years. The top sources of grain foods were determined, and their contribution to thiamin, riboflavin, niacin, vitamin B6, and folic acid intakes were analyzed. Results: The top source of whole grains was whole wheat/rye bread for all ethnic-sex groups, followed by popcorn and cooked cereals, except for Native Hawaiian men and Japanese Americans, for whom brown/wild rice was the second top source; major contributors of refined grains were white rice and white bread, except for Latinos. Refined grain foods contributed more to grain consumption (27.1-55.6%) than whole grain foods (7.4-30.8%) among all ethnic-sex groups, except African American women. Grain foods made an important contribution to the intakes of thiamin (30.2-45.9%), riboflavin (23.1-29.2%), niacin (27.1-35.8%), vitamin B6 (22.9-27.5%), and folic acid (23.3-27.7%). Conclusions: This is the first study to document consumption of different grain sources and their contribution to B vitamins in five ethnic groups in the U.S. Findings can be used to assess unhealthful food choices, to guide dietary recommendations, and to help reduce risk of chronic diseases in these populations.

Content-aware compression for big textual data analysis

A substantial amount of information on the Internet is present in the form of text. The value of this semi-structured and unstructured data has been widely acknowledged, with consequent scientific and commercial exploitation. The ever-increasing data production, however, pushes data analytic platforms to their limit. This thesis proposes techniques for more efficient textual big data analysis suitable for the Hadoop analytic platform. This research explores the direct processing of compressed textual data. The focus is on developing novel compression methods with a number of desirable properties to support text-based big data analysis in distributed environments. The novel contributions of this work include the following. Firstly, a Content-aware Partial Compression (CaPC) scheme is developed. CaPC makes a distinction between informational and functional content in which only the informational content is compressed. Thus, the compressed data is made transparent to existing software libraries which often rely on functional content to work. Secondly, a context-free bit-oriented compression scheme (Approximated Huffman Compression) based on the Huffman algorithm is developed. This uses a hybrid data structure that allows pattern searching in compressed data in linear time. Thirdly, several modern compression schemes have been extended so that the compressed data can be safely split with respect to logical data records in distributed file systems. Furthermore, an innovative two layer compression architecture is used, in which each compression layer is appropriate for the corresponding stage of data processing. Peripheral libraries are developed that seamlessly link the proposed compression schemes to existing analytic platforms and computational frameworks, and also make the use of the compressed data transparent to developers. The compression schemes have been evaluated for a number of standard MapReduce analysis tasks using a collection of real-world datasets. In comparison with existing solutions, they have shown substantial improvement in performance and significant reduction in system resource requirements.