Understanding Data

2.1 Core concepts

• Data defined: Data is a collection of characters, numbers, and other symbols that represent values of some situations or variables. The singular of "data" is "datum". Data need to be gathered, processed, and analysed for making decisions. (NCERT §5.1, p. 82)
• Knowledge base: A knowledge base is a store of information consisting of facts, assumptions, and rules which an AI system can use for decision making. (NCERT §5.1, p. 82 sidebar)
• Importance of data: Large amounts of data, when processed with a computer, reveal possibilities or hidden traits not otherwise visible. Examples include ATM transactions, meteorological satellite monitoring, dynamic pricing by airlines and cab apps, and market analysis by businesses. (NCERT §5.1.1, p. 82–83)
• Examples of data: Name/age/gender/contact details; banking and ticketing transaction data; images, graphics, audio, video; documents and web pages; online posts and messages; signals from sensors; satellite and meteorological data. (NCERT §5.1, p. 82)
• Structured data: Data organised in a well-defined format, usually stored in tabular (rows and columns) format where each column is an attribute/characteristic/variable and each row is an observation. Example: inventory table with columns ModelNo, ProductName, UnitPrice, Discount(%), Items_in_Inventory. (NCERT §5.1.2(A), p. 84)
• Unstructured data: Data not in a fixed row-and-column structure. Examples include web pages, text documents, business reports, books, audio/video files, social media messages. Unstructured data are sometimes described with the help of metadata. (NCERT §5.1.2(B), p. 85)
• Metadata: Data about data. For an email: subject, recipient, main body, attachment. For an image file: image size (KB/MB), image type (JPEG, PNG), image resolution. (NCERT §5.1.2(B), p. 85)
• Data collection: Identifying already-available data or collecting from appropriate sources. Data may exist in a diary/register (needs digitising), already in a digital file such as CSV, or may need a new software system to record it. (NCERT §5.2, p. 85–86)
• CSV: Comma Separated Values — a digital format in which data can already be available and ready for use. (NCERT §5.2, p. 85)
• Data storage: The process of storing data on storage devices so that data can be retrieved later. Common storage devices include Hard Disk Drive (HDD), Solid State Drive (SSD), CD/DVD, Tape Drive, Pen Drive, Memory Card. File processing limitations can be overcome through DBMS. (NCERT §5.3, p. 86)
• Data processing: Raw data (numbers/text/image) is transformed through processing into information (tables/charts/text). The Data Process Cycle has three stages — Input (Data Collection, Data Preparation, Data Entry), Processing (Store, Retrieve, Classify, Update), Output (Reports, Results, Processing System). (NCERT §5.4, p. 87, Figure 5.1)
• Measures of central tendency: A single value that gives an idea about the data. The three most common measures are Mean, Median, and Mode. (NCERT §5.5.1, p. 88)
• Mean: Average of numeric values of an attribute. Formula: sum of all n values divided by n. Mean is not suitable when there are outliers in the data. (NCERT §5.5.1(A), p. 88)
• Outlier: An exceptionally large or small value compared to other values; usually considered an error that can influence average calculations. (NCERT §5.5.1(A), p. 88 note)
• Median: When all values are sorted in ascending or descending order, the middle value is the Median. For odd number of values it is the middle position value; for even number of values it is the average of the two middle values. Median represents the central value at which the given data is equally divided into two parts. (NCERT §5.5.1(B), p. 89)
• Mode: The value that appears most number of times in the given data. Computed on the basis of frequency of occurrence. A dataset has no mode if each value occurs only once. There may be multiple modes if more than one value shares the highest frequency. Mode can be found for numeric as well as non-numeric data. (NCERT §5.5.1(C), p. 89)
• Measures of variability (dispersion): Refer to the spread or variation of values around the mean. Two common measures are Range and Standard Deviation. Two data sets can have the same mean/median/mode but completely different levels of dispersion. (NCERT §5.5.2, p. 89)
• Range: Difference between the maximum and minimum values (M - S). Can be calculated only for numerical data. Badly influenced by outliers since it uses only two extreme values. (NCERT §5.5.2(A), p. 90)
• Standard deviation (σ): Positive square root of the average of the squared difference of each value from the mean. Considers all given data values (unlike Range). Smaller σ means less spread; larger σ means more spread. Formula: σ = sqrt(Σ(xi − x̄)² / n). (NCERT §5.5.2(B), p. 90)
• Python for data analysis: Python has libraries specially built for data processing and analysis. It is one programming tool used for efficient analysis of large volumes of data using statistical techniques. (NCERT §5.5, p. 91)
• Why averages can mislead (NCERT §5.5.1(A), p. 88). If one student in a class of 30 scores 100 and the rest score around 50, the mean rises significantly above 50 even though most students are around 50 — this is the classic outlier-distortion scenario. Switch to median in such cases. Mean is best when the data is symmetric and outlier-free.
• Mode for categorical data (NCERT §5.5.1(C), p. 89). The mode is the only one of the three measures of central tendency that meaningfully applies to categorical/non-numeric data, e.g., the most popular favourite colour in a class, the most viewed video category. Mean and median are inherently numeric.
• Range vs IQR (NCERT §5.5.2(A), p. 90). The NCERT only introduces range, not interquartile range; CUET sticks to NCERT scope. Range = max − min. Range can be zero when all values are identical (no spread).
• Standard deviation interpretation (NCERT §5.5.2(B), p. 90). Smaller σ means data points cluster around the mean; larger σ means high spread. σ uses every data point — that is why it is preferred over range as a measure of dispersion.
• Data processing examples (NCERT §5.4, Figure 5.2, p. 87). Three canonical cases: competitive-exam website (input = candidate details/test answers, output = score/rank report), bank ATM withdrawal (input = card + PIN + amount, output = receipt + cash), train ticket issue (input = passenger details/source/destination, output = ticket + seat assignment). These illustrate the IPO cycle.

2.2 Definitions to memorise

2.3 Diagrams / processes to remember

• Figure 5.1 — Steps in Data Processing (p. 87): Shows two diagrams. First: RAW DATA (Numbers/Text/Image) → Data Processing → INFORMATION (In the form of table/chart/text). Second: Data Process Cycle with Input (Data Collection, Data Preparation, Data Entry) → Processing (Store, Retrieve, Classify, Update) → Output (Reports, Results, Processing System).
• Figure 5.2 — Data Based Problem Statements (p. 87): Three real-world scenarios (competitive exam website, bank ATM withdrawal, train ticket issue) each broken into Problem Statement, Inputs, Processing steps, and Output. Useful for identifying input-processing-output in application questions.
• Table 5.1 — Structured data about kitchen items in a shop (p. 84): Columns are ModelNo, ProductName, UnitPrice, Discount(%), Items_in_Inventory. Illustrates the attribute (column) and observation (row) structure of structured data.
• Table 5.3 — Standard deviation of attendance of 9 students (p. 91): Step-by-step calculation of σ for the height dataset [90, 102, 110, 115, 85, 90, 100, 110, 110] giving σ = 10.2 cm. Memorise the procedural steps: subtract mean from each value, square, sum, divide by n, take square root.

2.4 Common confusions / NTA trap points

• Mean vs. Median for outliers: Mean is sensitive to outliers (one extreme value distorts it); Median is not. NTA often presents a dataset with an extreme value and asks which measure is more appropriate — answer is Median.
• Range vs. Standard deviation: Both measure variability, but Range uses only two values (max and min) and is therefore badly affected by a single outlier. Standard deviation uses all values. A trap question asks which is a "better" or "more reliable" measure of dispersion — standard deviation is preferred.
• Mode for non-numeric data: Mean and Range can be calculated only for numeric data; Mode can be applied to both numeric and non-numeric data. A favourite NTA trap is to claim Mode is only for numbers.
• Structured vs. Unstructured misclassification: Students often label email body as structured data. The body is unstructured; the metadata (subject, recipient, attachment) gives it partial structure. Newspaper layout, tweets, audio files are all unstructured.
• No mode vs. multiple modes (NCERT §5.5.1(C), p. 89). A dataset where every value is unique has no mode. If two or more values share the highest frequency, the dataset has multiple modes. NTA may assert "a dataset always has exactly one mode" — this is false.
• Mean has no necessary connection to any actual data point (NCERT §5.5.1(A), p. 88). The mean might not appear anywhere in the dataset. NTA distractor: claims mean must be one of the observations.
• Range only on numerical data (NCERT §5.5.2(A), p. 90). Cannot compute range on categorical data. NTA tests this nuance.
• CSV is just a file format, not a data type (NCERT §5.2, p. 85). It is a structured representation but the values inside may be numbers or strings.
• Information ≠ Data (NCERT §5.4, p. 87). Information is the output of processing data — they are not interchangeable.
• DBMS overcomes file processing limitations (NCERT §5.3, p. 86). Flat files have redundancy/consistency issues; DBMS resolves them.
• Median splits the data 50/50 (NCERT §5.5.1(B), p. 89). Half of all values are ≤ median and half ≥ median.