TRACE: A Dynamic Model of Trust for People-Driven Service Engagements

Transcription

1 TRACE: A Dynamic Model of Trust for People-Driven Service Engagements Combining Trust with Risk, Commitments, and Emotions Anup K. Kalia Advisor: Munindar P. Singh Department of Computer Science North Carolina State University Raleigh, NC 27695, USA September 30, 2015 Anup Kalia (NCSU) TRACE September 30, / 24

2 Broader Objectives Understand subtle human and organizational relationships Use such relationships as a basis for estimating trust Anup Kalia (NCSU) TRACE September 30, / 24

3 Research Question How to estimate trust between people from their interactions? Possible Applications Support people to make important decisions in organizational settings Estimating team cohesion or performance Anup Kalia (NCSU) TRACE September 30, / 24

4 Limitations With Existing Approaches Several approaches consider commitments alone for trust estimation. Gambetta (1988) interprets trust as a truster s assessment of a trustee for performing a specific task Mayer et al. (1995) define trust as the willingness of a truster to be vulnerable to a trustee for the completion of a task Teacy et al. (2006) consider trust as the truster s estimation of probability that a truster will fulfill it s obligation toward a trustee Wang et al. (2011) represent trust as the belief of a truster that trustee will cooperate. They estimate trust by aggregating positive and negative experiences Kalia et al. (2014) consider commitment outcomes to predict trust where they learn truster s parameters based on whether outcomes are positive, negative, or neutral Anup Kalia (NCSU) TRACE September 30, / 24

5 Limitations With Existing Approaches Two major classes of trust models Fixed parameter trust models where parameter are manually fixed Machine-learned trust models typically Hidden Markov Models (HMM) that assumes variables are conditionally independent of each other given the output variable Anup Kalia (NCSU) TRACE September 30, / 24

6 Proposed Approach We can improve trust prediction by incorporating (in addition to commitments) two attributes Risk taken by a truster toward a trustee Risk taken depends on a truster s belief about the likelihood of gains or losses it might incur from its relationships with a trustee Emotions displayed by a truster toward a trustee Studies in psychology suggest that positive emotions increase trust whereas negative emotions decrease trust Create TRACE a model based on Conditional Random Field (CRF) Conditional independences between risk, commitments, and emotions may not hold in our setting (e.g., in HMM) Anup Kalia (NCSU) TRACE September 30, / 24

7 Background: Commitment & Trust Anup Kalia (NCSU) TRACE September 30, / 24

8 Background: Commitment Lifecycle C(Debtor, Creditor, Antecedent, Consequent) create conditional null expire detached cancel terminated consequent discharged cancel violated Anup Kalia (NCSU) TRACE September 30, / 24

9 Background: Estimating Trust from Commitment Progression Two-valued representation, positive and negative experiences: r, s Trust α = r r+s We characterize each subject via four parameters Initial values, r in, s in Increment for positive and negative experiences: i r and i s Commitment Operation Trust r fi, s fi Commissive create Directive create Delegate None λi r + r in, (1-λ)i s + s in Discharge i r + r in, s in Cancel r in, i s + s in Anup Kalia (NCSU) TRACE September 30, / 24

10 Trust Antecedent Framework (Mayer et al., 1995) We propose TRACE based on the enhance trust antecedent framework The model contains 4 variables trust (T), risk (R), commitments (C), and emotions (E) Each variable V = T, R, C, E is described using using Singh s (1999, 2011) formal notation V debtor, creditor, antecedent, consequent Anup Kalia (NCSU) TRACE September 30, / 24

11 Description of Variables C trustee, truster, antecedent, consequent The trustee commits to the truster to perform the consequent If the trustee performs the consequent, the commitment is satisfied R truster, trustee, antecedent, consequent The truster takes a risk by accepting the trustee s offer to perform the consequent If the trustee performs the consequent, the truster gains T truster, trustee, antecedent, consequent The truster believes the trustee if the trustee performs the consequent Trust has three dimensions: ability (trustee s competency), benevolence (trustee s willingness), integrity (trustee s ethics and morality) E truster, trustee, antecedent, consequent The truster displays a positive emotion if the trustee performs the consequent Anup Kalia (NCSU) TRACE September 30, / 24

12 Postulates We propose postulates that capture relationships between the variables P 1 P 2 P 3 : T t T t+1. The trust T t+1 is influenced by the past trust T t : C t T t. The current commitment outcome C t influences the current trust T t : R t C t. The risk taken influences the commitment outcome C t or the gain or loss realized in the risk R t P 4 P 5 P 6 P 7 : R t T t. The current risk taken R t influences the current trust T t : C t E t. The commitment outcome C t influences the current emotion E t : R t E t. The risk taken R t influences the truster s emotion E t : E t T t. The current emotion E t influences the current trust T t Anup Kalia (NCSU) TRACE September 30, / 24

13 The TRACE Model Graphical representation of HMM and TRACE trust models (two time slices) ld ld ld ld ld ld ld ld Anup Kalia (NCSU) TRACE September 30, / 24

14 Comparing HMM and CRF HMM makes two independent assumptions CRF The current state y t is independent of y 1, y 2,..., y t 2, given y t 1 Observations x t are independent of each other, given y t CRFs are agnostic to dependencies between the observations CRF model employs discriminative modeling, where the distribution p( y x) is learned directly from the data Anup Kalia (NCSU) TRACE September 30, / 24

15 Evaluation We evaluate TRACE via data collected from a human-subject study conducted by the Intelligence Advanced Research Projects Activity (IARPA) IARPA prepared a dataset based on the Checkmate protocol adapted from the investment or dictator economic decision-making game (Berg, 1995) The data consists of 431 rows collected from 63 subjects Each row corresponds to the sequence of rounds played between two subjects The data we obtained reflects only the banker s perspective W d Z W d Z W d Z W d Z W d Z W d Z W d Z W d Z ' W ' W ' W ' W ' W ' W ' W ' W Anup Kalia (NCSU) TRACE September 30, / 24

16 Evaluation The protocol involves two roles: banker and game player The banker s task is to loan money to a game player The game player requests a loan from the banker to play a maze, promising to play a maze of certain difficulty and return: the loan with all gains, the loan with 50% of all gains, 50% of the available money, a fixed amount After the game player s request, the banker chooses a loan category: small (1 7 USD), medium (4 10 USD), or big (7 13 USD) A dollar amount, randomly generated within the banker s chosen category, is loaned to the game player The game player does not know the category chosen by the banker The game player plays a maze of a certain difficulty (not necessarily what he or she had promised) The banker will not know the actual maze played The game player returns some money to the banker (not necessarily what he or she had promised) Anup Kalia (NCSU) TRACE September 30, / 24

17 Mapping Game Elements to The TRACE Model The commitment from the player to the banker, C player,banker,loan,return, as satisfied if the player returned at least the amount he or she had loaned, and as violated, otherwise We compute the gain or loss in the risk, R banker, player, loan, return, based on the difference between the loaned and returned amounts The dataset represents the banker s trust for the player after the round, T banker,player,loan,return, as a three-tuple A, B, I, indicating the banker s perception of player s ability, benevolence, and integrity The dataset represents the banker s emotion after he or she receives a return from the player, E banker,player,loan,return, as real-valued (1 10) state anxiety scores derived from the post-round questionnaire Anup Kalia (NCSU) TRACE September 30, / 24

18 Results MAEs of HMM and TRACE considering different feature combinations. Model Input Variables A B I HMM TRACE C C + R C + E R + E C + R + E C C + R C + E R + E C + R + E Anup Kalia (NCSU) TRACE September 30, / 24

19 Results Considering only C, TRACE yields lower MAEs than HMM for each trust attribute (CRF employs discriminative) Considering all features (C + R + E), TRACE again yields lower MAEs than HMM for each trust attribute (CRF captures dependencies between C, R, and E) Considering C and R, TRACE performs better than HMM in predicting A and I (C and R are not independent) Considering C and E, HMM performs better than TRACE for B and I (C and E are conditionally independent) Considering R and E, TRACE performs better than HMM for B and I whereas HMM performs better than TRACE for A (mixed) Anup Kalia (NCSU) TRACE September 30, / 24

20 Threats to Validity Our dataset, although real, consists of short sequences. We expect both HMM and TRACE to perform better given longer sequences The dataset is skewed toward positive trust values and our conclusions may not hold since the trust values have a different distribution The dataset represents emotions using anxiety scores only, thereby lacking realistic emotion responses along multiple dimensions such as anger and joy Anup Kalia (NCSU) TRACE September 30, / 24

21 Discussion TRACE illustrates that a probabilistic model of trust that incorporates commitments, risk, and emotions can produce trust estimates with fairly good accuracy Our findings therefore open up the possibility of developing user agents that promote secure collaboration Using TRACE a user can calibrate the perceived trust with the risk undertaken in light of available measures of risk and gain from commitments Anup Kalia (NCSU) TRACE September 30, / 24

22 Thanks Anup Kalia (NCSU) TRACE September 30, / 24

23 Background: Identifying Commitment Operations from Interactions Ten-fold cross-validation using SVM on marked up Enron sentences Commitment Operation Precision Recall F-measure Commissive create Directive create Delegate Discharge Cancel Features used in the classifier include (out of 15) 1 Modal verb (shall, will, may, might, can, could, would, must) 2 Type of subject (first person, second person, third person) 3 Present tense verb 4 Past tense verb 5 Deadline Anup Kalia (NCSU) TRACE September 30, / 24

24 Background: Determining Commitment Operations and Trust from Text Commitments being the most prominent normative relationship S R Content Operation T S,R T R,S Kim Dorothy I will also check with Alliance Travel Agency... Kim Dorothy I checked with our Travel Agency... Rob Kim By Wednesday Aug , please send all copies of your documentation... Kim Rob Rob, please forgive me for not sending this in by Aug 15 create(c 1 ) discharge(c 1 ) create(c 2 ) cancel(c 2 ) Example s from the Enron corpus Anup Kalia (NCSU) TRACE September 30, / 24