An Activation Method of Topic Dictionary to Expand Training Data for Trend Rule Discovery

Algorithm 2

Pseudocode for prediction of attractive evaluation objects.

(01) For 𝑐∈𝐶

(02)  𝑅[𝑐] = pickUpTrendRule𝑅[𝑐],𝑀_𝑖;

(03) initializeEvaluationObjectCounter(𝑀);

(04) For 𝑑∈𝐷

(05)  𝑊_𝑑 = extractEvaluationObjectAndAttribute(𝑑);

(06)  initializeClassCounter(𝑁);

(07)  For 𝑐∈𝐶

(08)   For 𝑟∈𝑅[𝑐]

(09)    𝑓 = evaluateTrendRule𝑊_𝑑,𝑟;

(10)    if 𝑓==TRUE

(11)       𝑁[𝑐] += 1;

(12)  𝑐_𝑑 = identifyClassFor Element(𝑁);

(13)  if 𝑐_𝑑≠𝜙

(14)   For 𝑤∈𝑊_𝑑

(15)    if 𝑤∈𝐸

(16)     𝑀𝑤,𝑐_𝑑 += 1;

(17) For 𝑒∈𝐸

(18)  𝑓 = identifyAttractiveEvaluationObject𝑀[𝑒,⋅],𝑀_𝑡;

(19)  if 𝑓==TRUE

(20)   output𝑒,𝑐_𝑒,𝑀𝑒,𝑐_𝑒;