問19
頻出度★★★
ビッグデータの分析に関する記述として, 最も適切なものはどれか。
令和3年度 ITパスポート試験 問19
ストラテジ系企業活動 › 業務分析・データ利活用
This question's explanation and your answers so far.
解説
問 19
- ア
正解 In big data analysis, a major goal is to predict as-yet-unknown trends and patterns, as well as future situations, from the large volumes of diverse data that have been accumulated. Here, by combining not only statistical analysis methods such as averages, correlation, and regression, but also machine learning, which automatically learns patterns from data to build models, unknown situations can be predicted with higher accuracy. The two are not competing but complementary, so the description in choice ア that machine learning is also effective in addition to statistical methods is the most appropriate.
Incorrectイ
Incorrect. One of the characteristics of big data is Variety; not only text but also unstructured data such as video, images, and audio can be objects of analysis. Analysis of such non-text data, including image recognition and speech recognition, is in fact widely carried out, so saying it cannot be handled is contrary to the facts.
Incorrectウ
Incorrect. Not only information generated by humans (SNS, bulletin boards, search histories) but also data obtained from machines or transactions, such as sensors and IoT devices, purchase histories, and location information, is used to analyze human behavior. Limiting it to only information generated by humans is overly narrow and inappropriate.
Incorrectエ
Incorrect. Even data that was not originally intended to be analyzed, such as blog posts, can yield useful insights through methods like text mining. Uniformly excluding such data from the targets regardless of the purpose of analysis runs counter to the idea of leveraging big data.
ここがポイント
- In big data analysis, statistical methods and machine learning are combined to predict unknown situations from large volumes of data (the two are complementary).
- Big data has variety; not only text but also unstructured data such as video, images, and audio, as well as data originating from non-human sources (sensors, IoT), are objects of analysis.
用語