IF-ENTROPY: Intuitionistic Fuzzy Entropy Weight Method (Vlachos-Sergiadis 2007 entropy measure as applied by Hung-Chen 2010)

Overview

Information-theoretic objective weighting under Intuitionistic Fuzzy uncertainty (IF entropy → divergence → simplex-normalised crisp weights). Output typically weight (higher value = preferred).

Strengths

• •Method-specific: Information-theoretic objective weighting under Intuitionistic Fuzzy uncertainty (IF entropy → divergence → simplex-normalised crisp weights)
• •Preserves intuitionistic uncertainty through the pipeline rather than premature crispification at elicitation

Limitations

• •Assumes: Decision matrix entries are valid IFNs (μ+ν ≤ 1)
• •Assumes: At least one criterion has non-uniform IFN entries across alternatives (otherwise weights undefined)
• •Assumes: If multi-DM: DM weights λ_k crisp and sum to 1 (l-simplex)

Assumptions to verify

• •Decision matrix entries are valid IFNs (μ+ν ≤ 1)
• •At least one criterion has non-uniform IFN entries across alternatives (otherwise weights undefined)
• •If multi-DM: DM weights λ_k crisp and sum to 1 (l-simplex)

When not to use

• •Crisp data sufficient: use base ENTROPY (Shannon 1948) directly
• •DM has strong subjective preferences: use AHP, BWM, or SWARA instead
• •All criteria are highly uniform across alternatives: IF-ENTROPY cannot discriminate, weights degenerate

Edge cases

• •If multi-DM input is supplied, also verify DM weights λ on the l-simplex (Σλ=1). Construct the IF decision matrix (or k matrices for group decision making).
• •IF decision matrix D̃ = (r̃_ij) using the Xu 2007 IFWA operator with DM weights λ. For single-DM input (dm_count=1) this step is the identity.
• •IF entropy per criterion E^IFS_LT(C_j) using the Vlachos-Sergiadis 2007 measure (Hung-Chen 2010 Eq. 7). For each criterion j: sum over alternatives i of the four-term integrand [μ ln μ + ν ln ν − (1−π

Common pitfalls

• •Hatalı: 'IF-ENTROPY bu varsayımı kontrol etmeden uygulamak'. Doğrusu: Decision matrix entries are valid IFNs (μ+ν ≤ 1)
• •Hatalı: 'IF-ENTROPY bu varsayımı kontrol etmeden uygulamak'. Doğrusu: At least one criterion has non-uniform IFN entries across alternatives (otherwise weights undefined)
• •Hatalı: 'IF-ENTROPY bu varsayımı kontrol etmeden uygulamak'. Doğrusu: If multi-DM: DM weights λ_k crisp and sum to 1 (l-simplex)
• •Hatalı: IF-ENTROPY'yi 'Crisp data sufficient' durumunda kullanmak: recommendation_metadata.not_recommended_when alternatif öneriyor.
• •Hatalı: IF-ENTROPY'yi 'DM has strong subjective preferences' durumunda kullanmak: recommendation_metadata.not_recommended_when alternatif öneriyor.
• •Hatalı: IF-ENTROPY'yi 'All criteria are highly uniform across alternatives' durumunda kullanmak: recommendation_metadata.not_recommended_when alternatif öneriyor.

Worked example

1. 1.Adım 1 (F1): Step 1: Validate IFN axioms (μ+ν ≤ 1) on every matrix entry. If multi-DM input is supplied, also verify DM weights λ on the l-simplex (Σλ=1). Construct the IF decision matrix (or k matrices for group decision making). Formül: R^{(k)} = (r^{(k)}_{ij})_{m \times n},\ r^{(k)}_{ij} = (\mu^{(k)}_{ij}, \nu^{(k)}_{ij}),\ \mu+\nu \le 1,\ \sum_k \lambda_k = 1 Anchor: Atanassov 1986 (IFS axioms); Hung-Chen 2010 Step 1 (Eqs. 4-5)
2. 2.Adım 2 (F2): Step 2: (Multi-DM only; identity for single-DM) Aggregate the k individual DM matrices R^(1), ..., R^(k) into a single collective IF decision matrix D̃ = (r̃_ij) using the Xu 2007 IFWA operator with DM weights λ. For single-DM input (dm_count=1) this step is the identity. Formül: \tilde{r}_{ij} = \text{IFWA}_{\lambda}(r^{(1)}_{ij}, \ldots, r^{(k)}_{ij}) = \left(1 - \prod_{k} (1 - \mu^{(k)}_{ij})^{\lambda_k},\ \prod_{k} (\nu^{(k)}_{ij})^{\lambda_k}\right) Anchor: Hung-Chen 2010 Eq. (6); Xu 2007 IFWA operator (Def 3.3, Eq. 12)
3. 3.Adım 3 (F3): Step 3: Compute the IF entropy per criterion E^IFS_LT(C_j) using the Vlachos-Sergiadis 2007 measure (Hung-Chen 2010 Eq. 7). For each criterion j: sum over alternatives i of the four-term integrand [μ ln μ + ν ln ν − (1−π) ln(1−π) − π ln 2], scaled by −1/(m ln 2) so that E ∈ [0, 1]. Convention: x ln x = 0 when x = 0 (limit as x→0⁺). Formül: E^{\text{IFS}}_{LT}(C_j) = -\frac{1}{m \ln 2} \sum_{i=1}^{m} \left[ \mu_{ij} \ln \mu_{ij} + \nu_{ij} \ln \nu_{ij} - (1 - \pi_{ij}) \ln(1 - \pi_{ij}) - \pi_{ij} \ln 2 \right] Anchor: Hung-Chen 2010 Eq. (7); Vlachos-Sergiadis 2007 (entropy measure source); Szmidt-Kacprzyk 2001 (four axioms)
4. 4.Adım 4 (F4): Step 4: Compute the divergence (information content) per criterion d_j = 1 − E^IFS(C_j). Higher d_j means lower entropy, hence higher information content and discrimination power. Formül: d_j = 1 - E^{\text{IFS}}_{LT}(C_j),\ j = 1, \ldots, n Anchor: Hung-Chen 2010 Eq. (8)
5. 5.Adım 5 (F5): Step 5: Normalise the divergence vector to crisp weights on the n-simplex: w_j = d_j / Σ_k d_k. The output is a crisp weight vector w ∈ Δ^{n-1} satisfying Σw_j = 1, w_j ≥ 0. Formül: w_j = \frac{d_j}{\sum_{k=1}^{n} d_k},\ \sum_j w_j = 1,\ w_j \ge 0 Anchor: Hung-Chen 2010 Eq. (9)

Commonly paired with

• •IF-ENTROPY + IF-TOPSIS (common)
• •IF-ENTROPY + IF-MAUT (common)
• •IF-ENTROPY + IF-VIKOR (occasional)

How to cite