DFAs, NFAs, and regular languages

For each statement below, state if it is true or false, and explain why. The explanation does not need to be a formal proof, but the argument should be sound.

Statement a

If $L_1$ is regular and $|L_1| = k$ and $L_2$ is non-regular, then $L_1 \cap L_2$ is regular.

This statement is false.

All finite languages are regular. $|L_1| = k$ implies that $L_1$ is finite, and therefore regular. The intersection of a regular language and a non-regular language is not guaranteed to be regular.

Note that all string under $L_1 \cap L_2$ must be a subset of $L_1$, and a subset of a finite language is finite, therefore regular.

For example, let $L_1$ be a regular language that contains a string $a^nb^n$ and $L_2 = \{a^nb^n\}$.

The intersection of $L_1 \cap L_2$ is non-regular.

Statement b

If $L_1$ and $L_2$ are non-regular, then $L_1 \cup L_2$ is regular.

This statement is false.

The union of a regular and a non-regular language is not guaranteed to be regular.

A language is regular if there is an finite automaton that accepts it.

Note that $L_1$ is a regular language, therefore finite, and $L_2$ is non-regular, therefore there does not exist a finite automaton that accepts it.

If $L_1 \cup L_2$ is regular, then there must exist a finite automaton that accepts it. However, such automaton would also accept $L_2$ since $L_2 \subseteq L_{1} \cup L_2$, therefore meet contradiction.

Which renders the statement false.

Statement c

$\forall L_1 \mid L_1 \text{ :non-regular, } \exists L_2 \mid L_2 \text{ :regular} \land L_{1} \subseteq L_{2}$

This statement is true.

Let $\Sigma$ be the alphabet of $L_1$., choose $L_2 = \Sigma^{*}$, which is regular.

Since $\Sigma^{*}$ is the set of all strings formed from $\Sigma$ plus empty string, it is guaranteed to contain $L_1$. Therefore $L_1 \subseteq \Sigma^{*}$. Therefore, $L_1 \subseteq L_2$

Create a DFA $M$ such that:

Statement a

M accepts all strings which begin with $b$ but do not contain the substring $bab$.

Statement b

$\mathcal{L}{(M)} = \lbrace a^ib^jc^k \mid i+j+k \text{ is a multiple of 3} \rbrace$, $\Sigma = \lbrace a,b,c \rbrace$

Statement c

$\mathcal{L}{(M)} = \lbrace x \mid \text{at least two a's in last three characters of x} \rbrace$

Via product construction, create a DFA $M$ such that

First create two machine: one where $n$ is a multiple and one where $m$ is a multiple of 3. Then create the “union” machine:

First, we will construct $M_1$:

Then, we will construct $M_2$:

From product construction, we will create $M$ based on $M_1$ and $M_2$:

Create an NFA which accepts all string in which the third last character is an $a$. Then via subset construction, create an equivalent DFA. Show all your work

Solution

We define the following NFA $(Q, \Sigma, \delta, q_0, F)$ with:

• $Q = \{q_0, q_1, q_2, q_3, q_4\}$
• $\Sigma = \{a, b\}$
• Start state $q_0$
• Accept state $q_{3}$
• Transition function $\delta$ as follows:

Via subset construction, we can create the following DFA:

Start state of DFA is $\{ q_0 \}$, as it is the epsilon closure of the start state of the NFA

Transition table:

The final state are any states that include $q_{3}$, which are $\{q_{0}, q_{1}, q_{2}, q_{3}\}$ and $\{q_{0}, q_{3}\}$.

Where

dfa_states = {
    'D0': '{q0}',
    'D1': '{q0, q1}',
    'D2': '{q0, q2}',
    'D3': '{q0, q1, q2}',
    'D4': '{q0, q3}',
    'D5': '{q0, q1, q2, q3}',
    'D6': '{q4}'
}