Latent Tree ModelsPart IV: Applications

Latent Tree ModelsPart IV: Applications

AAAI 2014 Tutorial Latent Tree Models Part IV: Applications Nevin L. Zhang Dept. of Computer Science & Engineering The Hong Kong Univ. of Sci. & Tech. http://www.cse.ust.hk/~lzhang Applications of Latent Tree Analysis (LTA) What can LTA be used for: Discovery of co-occurrence patterns in binary data Discovery of correlation patterns in general discrete data Discovery of latent variable/structures Multidimensional clustering Topic detection in text data Probabilistic modelling Applications

Analysis of survey data Analysis of text data Market survey data, social survey, medical survey data Topic detection Approximate probabilistic inference AAAI 2014 Tutorial Nevin L. Zhang HKUST 2 Part IV: Applications Approximate Inference in Bayesian Networks Analysis of social survey data Topic detection in text data

Analysis of medical symptom survey data Software AAAI 2014 Tutorial Nevin L. Zhang HKUST 3 LTMs for Probabilistic Modelling Attractive Representation of Joint Distributions Computationally very simple to work with. Represent complex relationships among observed variables. What does the structure look like without the latent variables? AAAI 2014 Tutorial Nevin L. Zhang HKUST 4

Approximate Inference in Bayesian Networks In a Bayesian network over observed variables, exact inference can be computationally prohibitive. Two-phase approximate inference: Offline (Wang et al. AAAI 2008) Sample data set from the original network Learn a latent tree model (secondary representation) Online Make inference using the latent tree model. (Fast) Sample Learn LTM AAAI 2014 Tutorial Nevin L. Zhang HKUST

5 Empirical Evaluations Alternatives Original networks LTM (1k), LTM (10k), LTM (100k): with different sample size for Phase 1. CL (100k): Phase 1 learns Chow-Liu tree LCM (100k): Phase 1 learns latent class model Loopy Belief Propagation (LBP) ALARM, INSURANCE, MILDEW, BARLEY, etc. Evaluation: 500 random queries

Quality of approximation measured using KL from exact answer. AAAI 2014 Tutorial Nevin L. Zhang HKUST 6 Empirical Results C: cardinality of latent variables When C is large enough, LTM achieves good approximation in all cases. Better than LBP on g, d,h Better than CL on d, h. Key Advantage: Online phase is 2 to 3 orders of magnitude faster

than exact inference sparse AAAI 2014 Tutorial Nevin L. Zhang HKUST dense 7 Part III: Applications Approximate Inference in Bayesian networks Analysis of social survey data Topic detection Analysis of medical symptom survey data Software AAAI 2014 Tutorial Nevin L. Zhang

HKUST 8 Social Survey Data // Survey on corruption in Hong Kong and performance of the anti-corruption agency -- ICAC //31 questions, 1200 samples C_City: C_Gov: C_Bus: s0 s1 s2 s3 // very common, quite common, uncommon, very uncommon s0 s1 s2 s3 s0 s1 s2 s3 Tolerance_C_Gov: s0 s1 s2 s3 Tolerance_C_Bus: s0 s1 s2 s3 //totally intolerable, intolerable, tolerable, totally tolerable WillingReport_C: s0 s1 s2 // yes, no, depends LeaveContactInfo: s0 s1

I_EncourageReport: s0 s1 s2 s3 s4 // very sufficient, sufficient, average, ... I_Effectiveness: s0 s1 s2 s3 s4 //very e, e, a, in-e, very in-e I_Deterrence: s0 s1 s2 s3 s4 // yes, no // very sufficient, sufficient, average, ... .. -1 -1 -1 0 0 -1 -1 -1 -1 -1 -1 0 -1 -1 -1 0 1 1 -1 -1 2 0 2 2 1 3 1 1 4 1 0 1.0 -1 -1 -1 0 0 -1 -1 1 1 -1 -1 0 0 -1 1 -1 1 3 2 2 0 0 0 2 1 2 0 0 2 1 0 1.0 -1 -1 -1 0 0 -1 -1 2 1 2 0 0 0 2 -1 -1 1 1 1 0 2 0 1 2 -1 2 0 1 2 1 0 1.0 . AAAI 2014 Tutorial Nevin L. Zhang HKUST 9

Latent Structure Discovery Y2: Demographic info; Y3: Tolerance toward corruption; Y4: ICAC performance; Y5: Change in level of corruption; Y6: Level of corruption; Y7: ICAC accountability AAAI 2014 Tutorial Nevin L. Zhang HKUST 10 Multidimensional Clustering Y2=s0: Low income youngsters; Y2=s1: Women with no/low income; Y2=s2: people with good education and good income; Y2=s3: people with poor education and average income. AAAI 2014 Tutorial Nevin L. Zhang HKUST 11 Multidimensional Clustering

Y3=s0: people who find corruption totally intolerable; 57% Y3=s1: people who find corruption intolerable; 27% Y3=s2: people who find corruption tolerable; 15% Interesting finding: Y3=s2: 29+19=48% find C-Gov totally intolerable or intolerable; 5% for C-Bus Y3=s1: 54% find C-Gov totally intolerable; 2% for C-Bus Y3=s0: Same attitude toward C-Gov and C-Bus People who are tough on corruption are equally tough toward C-Gov and C-Bus. People who are lenient about corruption are more lenient C-Bus than C-GOv AAAI 2014 Tutorial Nevin L. Zhang HKUST 12 Multidimensional Clustering Who are the toughest toward corruption among the 4 groups? Y2=s2: ( good education and good income) the least tolerant. 4% tolerable Y2=s3: (poor education and average income) the most tolerant. 32% tolerable The other two classes are in between. Summary: Latent tree analysis of social survey data can reveal Interesting latent structures Interesting clusters Interesting relationships among the clusters. AAAI 2014 Tutorial Nevin L. Zhang HKUST 13 Part III: Applications

Approximate Inference Analysis of social survey data Topic detection (Analysis of text data) Analysis of medical symptom survey data Software AAAI 2014 Tutorial Nevin L. Zhang HKUST 14 Latent Tree Models for Topic Detection Basics Aggregation of miniature topics

Topic extraction and characterization Empirical results AAAI 2014 Tutorial Nevin L. Zhang HKUST 15 What is a topic in LTA? LTM for toy text data Topic: State of latent variable, soft collection of documents Characterized by: Conditional probability of word given latent state, or, document frequency of word in collection: # docs containing the word / total # of docs in the topic Probabilities all words for a topic (in a column) do not sum to 1. Y1=2: oop; Y1=1: Programming; Y1=0: background

Background topics for other latent variables not shown. AAAI 2014 Tutorial Nevin L. Zhang HKUST 16 How are topics and documents are related? Topic: A collection of documents A document is a member of a topic Can belong to multiple topics with different probabilities Probabilities for each document (in each row) do not sum to 1. D97, D115, D205, D528 are documents from the toy text data Table shows: D97 is a web page on OOP from U of Wisconsin Madison D528 is a web page on AI from U of Texas Austin AAAI 2014 Tutorial Nevin L. Zhang HKUST

17 LTA Differs from Latent Dirichlet Allocation (LDA) LDA Topic: Distribution over vocabulary Frequencies a writer would use each word when writing about the topic Probabilities for a topic (in a column) sum to 1 In LDA a document is a mixture of topics (LTA: Topic is a collection of documents) Probabilities in each row sum to 1 AAAI 2014 Tutorial Nevin L. Zhang HKUST 18 Latent Tree Models for Topic Detection Basics

Aggregation of miniature topics Topic extraction and characterization Empirical results AAAI 2014 Tutorial Nevin L. Zhang HKUST 19 Latent Tree Model for a Subset of Newsgroup Data Latent variable give miniature topics. Intuitively, more interesting topics can be detected if we combine Z11, Z12, Z13

Z14, Z15, Z16 Z17, Z18, Z19 BI algorithm produces flat models: Each latent variable directly connected to at least one observed variables. AAAI 2014 Tutorial Nevin L. Zhang HKUST 20 Hierarchical Latent Tree Analysis (HLTA) Convert the latent variables into observed one via hard assignment. Afterwards, Z11-Z19 become observed. Run BI on Z11-Z19 AAAI 2014 Tutorial Nevin L. Zhang HKUST 21 Hierarchical Latent Tree Analysis (HLTA)

Stack model for Z11-Z19 on top of model for the words Repeat until no more than 2 latent variables or predetermined level reached. The result is called a hierarchical latent tree model (HLTM) AAAI 2014 Tutorial Nevin L. Zhang HKUST 22 Hierarchical Latent Tree Analysis (HLTA) Part II: Cannot determine edge orientations based solely on data. Here hierarchical structure introduced to improve model interpretability. Data + interpretability hierarchical structure. It does not necessarily improve model fit.

AAAI 2014 Tutorial Nevin L. Zhang HKUST 23 Latent Tree Models for Topic Detection Basics Aggregation of miniature topics Topic extraction and characterization Empirical results AAAI 2014 Tutorial Nevin L. Zhang HKUST 24 Semantic Base Interpreting states of Z21

Z11, Z12, and Z13 introduced because of co-occurrence of computer, Science; card, display, ., video; and dos , windows Z21 introduced because of correlations among Z11, Z12, Z13 So, interpretation of the states of Z21 is to be based on the words in the sub-tree rooted at Z21. They form the semantic base of Z21. AAAI 2014 Tutorial Nevin L. Zhang HKUST 25 Effective Semantic Base Semantic base might be too large to handle.

Effective base: Subset of semantic base that matters. Sort variables Xi from semantic base in descending of I(Z; Xi). I(Z; X1, , Xi): Mutual information between Z and first i-th variables Chen et al. AIJ 2012 Estimated via sampling, increases with i. I(Z; X1, , Xm): Mutual information between Z and all m variables in semantic base Information coverage of the first i-th variable I(Z; X1, , Xi)/ I(Z; X1, , Xm): Effective semantic base:

Set of leading variables with information coverage higher than a certain level, i.e., 95%. AAAI 2014 Tutorial Nevin L. Zhang HKUST 26 Z22: Upper: Information coverage Lower: Mutual Information Effective semantic bases are typically smaller than Semantic bases. Z22: Semantic base --10 variables, Effective semantic base 8 variable Differences are much larger in models with hundreds of AAAI 2014 Tutorial Nevin L. Zhang variables. HKUST 27 Topic Characterizations

HLTA characterizes Latent state (topics) using probabilities of words from effective semantic base Topic Z22=s1 characterized using words NOT sorted according to probability, but mutual information Occur with high probabilities in documents on to the topic, and Occur with low probability in documents NOT on the topic. LDA, HLDA, Topic characterized using words that occur with highest probability in the topic. Not necessarily the best words to distinguish the topic from other topics. AAAI 2014 Tutorial Nevin L. Zhang HKUST 28

Latent Tree Models for Topic Detection Basics Aggregation of miniature topics Topic extraction and characterization Empirical results AAAI 2014 Tutorial Nevin L. Zhang HKUST 29 Empirical Results Show the results of HLTA on real-world data Compare HLTA with HLDA and LDA AAAI 2014 Tutorial Nevin L. Zhang

HKUST 30 NIPS Data 1,740 papers published at NIPS between 1988 1999. Vocabulary: HLTA produced a model with 382 latent variables, arranged on 5 levels. Level 1 279; Level 2 72; Level 3 - 21; Level 4 - 8; Level 5 - 2 Example topics on next few slides 1,000 words selected using average TF-IDF. Topic characterizations, topic sizes, Topic groups, topic group labels.

For details: http://www.cse.ust.hk/~lzhang/ltm/index.htm AAAI 2014 Tutorial Nevin L. Zhang HKUST 31 HLTA Topics: Level-3 likelihood bayesian statistical gaussian conditional reinforcement markov speech hmm transition 0.34 likelihood bayesian statistical conditional 0.16 gaussian covariance variance matrix 0.21 eigenvalues matrix gaussian covariance 0.20 markov speech speaker hmms hmm trained classification classifier regression classifiers 0.10 reinforcement sutton barto actions policy 0.25 validation regression svm machines 0.07 svm machines vapnik regression

0.38 trained test table train testing 0.30 classification classifier classifiers class cl 0.14 speech hmm speaker hmms markov 0.13 reinforcement sutton barto policy actions cells neurons cortex firing visual 0.17 visual cells cortical cortex activity 0.27 cells cortex cortical activity visual images image pixel pixels object 0.33 neurons neuron synaptic synapses 0.25 images image pixel pixels texture 0.16 receptive orientation objects object 0.21 object objects perception receptive 0.18 membrane potentials spike spikes firing 0.15 firing spike membrane spikes potentials 0.18 circuit voltage circuits vlsi chip hidden propagation layer backpropagation units

0.26 dynamics dynamical attractor stable attractors 0.40 hidden backpropagation multilayer architecture architectures 0.40 propagation layer units back net .. L. Zhang AAAI 2014 Tutorial Nevin HKUST 32 HLTA Topics: Level-2 markov speech hmm speaker hmms reinforcement sutton barto actions policy 0.14 markov stochastic hmms sequence hmm 0.12 transition states reinforcement reward 0.10 hmm hmms sequence markov

stochastic 0.10 reinforcement policy reward states 0.15 speech language word speaker acoustic 0.06 speech speaker acoustic word language 0.14 trajectory trajectories path adaptive 0.12 actions action control controller agent 0.09 sutton barto td critic moore 0.16 delay cycle oscillator frame sound 0.10 frame sound delay oscillator cycle 0.14 strings string length symbol AAAI 2014 Tutorial Nevin L. Zhang HKUST 33 HLTA Topics: Level-2 likelihood bayesian statistical conditional posterior 0.34 likelihood statistical conditional density 0.35 entropy variables divergence mutual

0.19 probabilistic bayesian prior posterior 0.11 bayesian posterior prior bayes 0.15 mixture mixtures experts latent 0.14 mixture mixtures experts hierarchical 0.34 estimate estimation estimating estimated 0.21 estimate estimation estimates estimated 0.24 regression svm vapnik margin kernel 0.05 svm vapnik margin kernel regression 0.19 validation cross stopping pruning 0.07 machines boosting machine boltzmann classification classifier classifiers class classes 0.28 classification classifier classifiers class 0.24 discriminant label labels discrimination gaussian covariance matrix variance eigenvalues 0.09 matrix pca gaussian covariance variance 0.23 gaussian covariance variance matrix pca 0.09 pca gaussian matrix covariance

variance 0.18 eigenvalues eigenvalue eigenvectors ij 0.15 blind mixing ica coefficients inverse regression validation vapnik svm machines 0.13 handwritten digit character digits trained test table train testing 0.38 trained test table train testing 0.44 experiments correct improved improvement correctly AAAI 2014 Tutorial Nevin L. Zhang HKUST 34 HLTA Topics: Level-1 likelihood statistical conditional density log 0.30 likelihood conditional log em maximum

mixture mixtures experts hierarchical latent 0.19 mixture mixtures 0.42 statistical statistics 0.34 multiple individual missing hierarchical 0.19 density densities 0.15 hierarchical sparse missing multiple 0.07 experts expert entropy variables variable divergence mutual 0.32 weighted sum 0.16 entropy divergence mutual 0.31 variables variable bayesian posterior probabilistic prior bayes 0.19 bayesian prior bayes posterior priors 0.09 bayesian posterior prior priors bayes 0.29 probabilistic distributions probabilities

0.16 inference gibbs sampling generative 0.19 mackay independent averaging ensemble 0.09 uk ac 0.38 estimate estimation estimated estimating 0.19 estimate estimates estimation estimated 0.29 estimator true unknown 0.33 sample samples 0.40 assumption assume assumptions assumed Reason for aggregate miniature 0.27 observations observation observed topics: Many Level 1 topics correspond to trivial word cooccurrences , not meaningful AAAI 2014 Tutorial 35 Nevin L. Zhang 0.08 belief graphical variational 0.09 monte carlo estimate estimation estimated estimates estimating

HKUST HLTA Topics: Level-4 & 5 Level 5 Level 4 visual cortex cells neurons firing 0.34 cells cortex firing neurons visual 0.28 cells neurons cortex firing visual visual cortex cells neurons firing 0.37 visual cortex firing neurons cells 0.41 approximation gradient optimization 0.39 visual cells firing cortex neurons 0.29 algorithms optimal approximation 0.25 images image pixel hidden trained 0.39 likelihood bayesian statistical gaussian 0.09 hidden trained images image pixel 0.20 trained hidden images image pixel

images image trained hidden pixel 0.15 image images pixel trained hidden 0.22 regression classification classifier 0.29 trained classification classifier classifiers 0.02 classification classifier regression 0.28 learn learned structure feature features 0.23 feature features structure learn learned 0.24 images image pixel pixels object 0.13 reinforcement transition markov speech 0.14 speech hmm markov transition 0.40 hidden propagation layer backpropagation units AAAI 2014 Tutorial Nevin L. Zhang HKUST 36 Summary of HLTA Results on NIPS Data Level 1: 279 latent variables Level 2: 72 latent variables

Meaningful topics, and meaningful topic groups More general than Level 2 topics Level 4: 8 latent variables Meaningful topics, very general Level 5: 2 latent variables Meaningful topics, and meaningful topic groups Level 3 : 21 latent variables Many capture trivial word co-occurrence patterns Too few In application, one can choose to output the topics at a certain level according the desired number of topics.

For NIPS data, either level-2 topics or level-3 topics. AAAI 2014 Tutorial Nevin L. Zhang HKUST 37 HLDA Topics units hidden layer unit weight gaussian log density likelihood estimate margin kernel support xi bound control optimal algorithms approximation step policy action reinforcement states actions experts mixture em expert gaussian convergence gradient batch descent means control controller nonlinear series forward distance tangent vectors euclidean distances robot reinforcement position control path bias variance regression learner exploration blocks block length basic experiment td evaluation features temporal expert path reward light stimuli paths Long hmms recurrent matrix term channel call cell channels rl generalization student weight teacher optimal gaussian bayesian kernel evidence posterior chip analog circuit neuron voltage

classifier rbf class classifiers classification speech recognition hmm context word ica independent separation source sources image images matching level object tree trees node nodes boosting variables variable bayesian conditional family face strategy differential functional weighting source grammar sequences polynomial regression derivative em machine annealing max min regression prediction selection criterion query validation obs generalization cross pruning image images recognition pixel feature video motion visual speech recognition face images faces recognition facial ocular dominance orientation cortical mlp risk classifier classification confidence loss song transfer bounds wt AAAI 2014 Tutorial principal curve eq curves rules cortex character characters pca coding field resolution false true detection context

. Nevin L. Zhang HKUST 38 LDA Topics inputs outputs trained produce actual dynamics dynamical stable attractor synaptic synapses inhibitory excitatory correlation power correlations cross units unit hidden connections connected states stochastic transition dynamic basis rbf radial gaussian centers solution constraints solutions constraint type elements group groups element edge light intensity edges contour recurrent language string symbol strings propagation back rumelhart bp hinton ii region regions iii chain experts expert gating architecture jordan hmm markov probabilities hidden hybrid object objects recognition view shape

robot environment goal grid world entropy natural statistical log statistics trajectory arm inverse trajectories hand sequence step sequences length s gaussian density covariance densities positive negative instance instances np target detection targets FALSE normal activity active module modules brain mixture likelihood em log maximum channel stage channels call routing graph matching annealing match term long scale factor range context mlp letter nn letters fig eq proposed fast proc variables variable belief conditional i AAAI 2014 Tutorial Nevin L. Zhang pp vol ca eds ieee HKUST 39 Comparisons between HLTA and HLDA HLTA Topics HLDA Topics likelihood bayesian statistical conditional posterior

gaussian log density likelihood estimate margin kernel support xi bound generalization student weight teacher optimal 0.34 likelihood statistical conditional density gaussian bayesian kernel evidence posterior chip analog circuit neuron voltage classifier rbf class classifiers classification 0.35 entropy variables divergence mutual 0.19 probabilistic bayesian prior posterior 0.11 bayesian posterior prior bayes speech recognition hmm context word 0.15 mixture mixtures experts latent 0.14 mixture mixtures experts hierarchical reinforcement sutton barto actions policy 0.12 transition states reinforcement reward 0.10 reinforcement policy reward states 0.14 trajectory trajectories path adaptive 0.12 actions action control controller agent 0.09 sutton barto td critic moore control optimal algorithms approximation

step policy action reinforcement states actions experts mixture em expert gaussian convergence gradient batch descent means control controller nonlinear series forward distance tangent vectors euclidean distances robot reinforcement position control path bias variance regression learner exploration blocks block length basic experiment HLTA topics have sizes, HLDA/LDA topics do not HLTA produces better hierarchy HLTA gives better topic characterizations AAAI 2014 Tutorial Nevin L. Zhang HKUST 40 Measure of Topic Quality Suppose a topic t is described using M words

The topic coherence score for t is: Idea The words for a topic would tend to co-occur. Given a list of words, the more often the words co-occur, than the better the list is as a definition of a topic. Note: Score decreases with M. Topics be compared should be described using the same number of words D. Mimno, H. M. Wallach, E. Talley, M. Leenders, and A. McCallum. Optimizing semantic coherence in topic models. In Proceedings of the Conference on Empirical Methods in Natural 20142011 Tutorial Language Processing, pages AAAI

262272, . Nevin L. Zhang HKUST 41 HLTA Found More Coherent Topics than LDA and HLDA HLTA (L3-L4): All non-background topics from Levels 3 and 4: 47 HLTA (L2-L3-L4): All non-background topics from Levels 2, 3 and 4: 140 LDA was instructed to find two sets of topics with 47 and140 topics HLDA found more 179. HLDA-s: A subset of the HLDA topics were sampled for fair comparison. AAAI 2014 Tutorial Nevin L. Zhang HKUST

42 Comparisons in Terms of Model Fit Regard LDA, HLDA and HLTA as methods for text modeling Evaluation: Build a probabilistic model for the corpus Per-document held-out loglikelihood (-log(perplexity)). Measure performance of model on predicting unseen data Data: NIPS: 1,740 papers from NIPS, 1,000 words, JACM: 536 abstracts from J of ACM, 1,809 words. NEWSGROUP: 20,000 newsgroup posts, 1,000 words. AAAI 2014 Tutorial Nevin L. Zhang

HKUST 43 HLTA results robust w.r.t UD-test threshold The values 1, 3, 5 are from literature on Bayes factor (see Part III) LDA produced by far worst models in all cases. HLTA out-performed HLDA on NIPS, tied on JACP, and beaten on Newsgroup Caution: Better model does not implies better topics Running time on NIPS: LDA 3.6 hours, HLTA 17 hours, HLDA 68 hours. AAAI 2014 Tutorial Nevin L. Zhang HKUST 44 Summary

HLTA Topic: collection of documents Have sizes Characterization: Words occur with high probability in topic, low probability in other documents Document: A member of topic, can belong to multiple topics with probability 1. LDA, HLDA Topic: Distribution over vocabulary

Dont have sizes Characterization: Words occur with high probability in topic Document: A mixture of topics HLTA produces better hierarchy than HLDA HLTA produce more coherent topics than LDA and HLDA AAAI 2014 Tutorial Nevin L. Zhang HKUST 45 Part III: Applications Approximate Inference in Bayesian networks Analysis of social survey data

Topic detection Analysis of medical symptom survey data Software AAAI 2014 Tutorial Nevin L. Zhang HKUST 46 Background of Research Common practice in China, increasingly in Western world Patients of a WM disease divided into several TCM classes Different classes are treated differently using TCM treatments. Example: WM disease: Depression TCM Classes: Liver-Qi Stagnation ( ). Treatment principle: Prescription:

Deficiency of Liver Yin and Kidney Yin ( ) Treatment principle: Prescription: Vacuity of both heart and spleen ( ). Treatment principle: , Prescription: Page 47 . AAAI 2014 Tutorial Nevin L. Zhang HKUST 47 Key Question How should patients of a WM disease be divided into subclasses from the TCM perspective? What TCM classes? What are the characteristics of each TCM class? How to differentiate different TCM classes?

Important for Clinic practice Research Randomized controlled trials for efficacy Modern biomedical understanding of TCM concepts No consensus. Different doctors/researchers use different schemes. Key weakness of TCM. Page 48 AAAI 2014 Tutorial Nevin L. Zhang HKUST 48 Key Idea Our objective: Provide an evidence-based method for TCM patient classification Key Idea Cluster analysis of symptom data => empirical partition of

patients Check to see whether it corresponds to TCM class concept Page 49 AAAI 2014 Tutorial Nevin L. Zhang HKUST 49 Symptoms Data of Depressive Patients Subjects: 604 depressive patients aged between 19 and 69 from 9 hospitals Selected using the Chinese classification of mental disorder clinic guideline CCMD-3 Exclusion: (Zhao et al. JACM 2014) Subjects we took anti-depression drugs within two weeks prior to the

survey; women in the gestational and suckling periods, .. etc Symptom variables From the TCM literature on depression between 1994 and 2004. Searched with the phrase and on the CNKI (China National Knowledge Infrastructure) data Kept only those on studies where patients were selected using the ICD-9, ICD-10, CCMD-2, or CCMD-3 guidelines. 143 symptoms reported in those studies altogether. Page 50 AAAI 2014 Tutorial Nevin L. Zhang HKUST 50 The Depression Data Data as a table 604 rows, each for a patient 143 columns, each for a symptom Table cells: 0 symptom not present, 1 symptom present

Removed: Symptoms occurring <10 times 86 symptoms variables entered latent tree analysis. Structure of the latent tree model obtained on the next two slides. Page 51 AAAI 2014 Tutorial Nevin L. Zhang HKUST 51 Model Obtained for a Depression Data (Top) Page 52 AAAI 2014 Tutorial Nevin L. Zhang HKUST 52 Model obtained for a Depression Data (Bottom) Page 53 AAAI 2014 Tutorial Nevin L. Zhang

HKUST 53 The Empirical Partitions The first cluster (Y29= s0) consists of 54% of the patients and while the cluster (Y29= s1) consists of 46% of the patients. The two symptoms fear of cold and cold limbs do not occur often in the first cluster While they both tend to occur with high probabilities (0.8 and 0.85) in the second cluster. Page 54 AAAI 2014 Tutorial Nevin L. Zhang HKUST 54 Probabilistic Symptom co-occurrence pattern Probabilistic symptom co-occurrence pattern:

The table indicates that the two symptoms fear of cold and cold limbs tend to co-occur in the cluster Y29= s1 Pattern meaningful from the TCM perspective. TCM asserts that YANG DEFICIENCY ( ) can lead to, among other symptoms, fear of cold and cold limbs So, the co-occurrence pattern suggests the TCM symdrome type YANG DEFICIENCY ( ). The partition Y29 suggests that Among depressive patients, there is a subclass of patient with YANG DEFICIENCY. In this subclass, fear of cold and cold limbs co-occur with high probabilities (0.8 and Page 55 AAAI 2014 Tutorial Nevin L. Zhang 0.85)

HKUST 55 Probabilistic Symptom co-occurrence pattern Y28= s1 captures the probabilistic co-occurrence of aching lumbus, lumbar pain like pressure and lumbar pain like warmth. This pattern is present in 27% of the patients. It suggests that Among depressive patients, there is a subclass that correspond to the TCM concept of KIDNEY DEPRIVED OF NOURISHMENT ( ) Characteristics of the subclass given by distributions for Y 28= s1 Page 56 AAAI 2014 Tutorial Nevin L. Zhang HKUST 56

Probabilistic Symptom co-occurrence pattern Y27= s1 captures the probabilistic co-occurrence of weak lumbus and knees and cumbersome limbs. This pattern is present in 44% of the patients It suggests that, Among depressive patients, there is a subclass that correspond to the TCM concept of KIDNEY DEFICIENCY Characteristics of the subclass given by distributions for Y27= s1 Y27, Y28, Y29 together provide evidence for defining KIDNEY YANG DEFICIENCY AAAI 2014 Tutorial Nevin L. Zhang HKUST 57 Probabilistic Symptom co-occurrence pattern

Pattern Y21= s1: evidence for defining STAGNANT QI TURNING INTO FIRE Y15= s1 : evidence for defining QI DEFICIENCY Y17 = s1 : evidence for defining HEART QI DEFICIENCY Y16= s1 : evidence for defining QI STAGNATION Y19= s1: evidence for defining QI STAGNATION IN HEAD Page 58 AAAI 2014 Tutorial Nevin L. Zhang HKUST 58 Probabilistic Symptom co-occurrence pattern Y9= s1 :evidence for defining DEFICIENCY OF BOTH QI AND YIN ( )

Y10= s1: evidence for defining YIN DEFICIENCY ( ) Y11= s1: evidence for defining DEFICIENCY OF STOMACH/SPLEEN YIN ( ) Page 59 AAAI 2014 Tutorial Nevin L. Zhang HKUST 59 Symptom Mutual-Exclusion Patterns Some empirical partitions reveal symptom exclusion patterns Y1 reveals the mutual exclusion of white tongue coating, yellow tongue coating and yellow-white tongue coating Y2 reveals the mutual exclusion of thin tongue coating, thick tongue

coating and little tongue coating. Page 60 AAAI 2014 Tutorial Nevin L. Zhang HKUST 60 Summary of TCM Data Analysis By analyzing 604 cases of depressive patient data using latent tree models we have discovered a host of probabilistic symptom cooccurrence patterns and symptom mutual-exclusion patterns. Most of the co-occurrence patterns have clear TCM syndrome connotations, while the mutual-exclusion patterns are also reasonable and meaningful. The patterns can be used as evidence for the task of defining TCM classes in the context of depressive patients and for differentiating between those classes. Page 61 AAAI 2014 Tutorial Nevin L. Zhang HKUST 61

(Zhang et al. JACM Another Perspective: Statistical Validation of TCM 2008) Postulates .. .. Kidney deprived of nourishment Y29 = s1 Yang Deficiency Y28 = s1 TCM terms such as Yang Deficiency were introduced to explain symptom co-occurrence patterns observed in clinic practice. Page 62 AAAI 2014 Tutorial Nevin L. Zhang HKUST 62 Value of Work in View of Others

D. Haughton and J. Haughton. Living Standards Analytics: Development through the Lens of Household Survey Data. Springer. 2012 Zhang et al. provide a very interesting application of latent class (tree) models to diagnoses in traditional Chinese medicine (TCM). The results tend to confirm known theories in Chinese traditional medicine. This is a significant advance, since the scientific bases for these theories are not known. The model proposed by the authors provides at least a statistical justification for them. Page 63 AAAI 2014 Tutorial Nevin L. Zhang HKUST 63 Part III: Applications

Approximate Inference in Bayesian networks Analysis of social survey data Topic detection Analysis of medical symptom survey data Software AAAI 2014 Tutorial Nevin L. Zhang HKUST 64 Software http://www.cse.ust.hk/faculty/lzhang/ltm/index.htm

Implementation of LTM learning algorithms: EAST, BI Tool for manipulate LTMs: Lantern LTM for topic detection: HLTA Implementation of other LTM learning algorithms BIN-A, BIN-G, CL and LCM: http://people.kyb.tuebingen.mpg.de/harmeling/code/ltt-1.4.tar CFHLC: https://sites.google.com/site/raphaelmouradeng/home/programs NJ, RG, CLRG and regCLRG: http://people.csail.mit.edu/myungjin/latentTree.html NJ (fast implementation): http://nimbletwist.com/software/ninja AAAI 2014 Tutorial Nevin L. Zhang HKUST 65

Recently Viewed Presentations

  • OpenMP for Cluster - Computer Science

    OpenMP for Cluster - Computer Science

    Irregular Accesses Indirect data accesses Data references are unknown at compile time Employ the Inspector-Executor strategy in our translation Study FIRE benchmark Gccg in FIRE Benchmark Inspector-Executor Inspector Collecting data access regions and save them into a hash table Separate...
  • The Earnings of Workers Covered by Pay Review Bodies ...

    The Earnings of Workers Covered by Pay Review Bodies ...

    The Role and Impact of the Pay Review Bodies Two Main Questions Outline of Talk 1. What are PRBs?? Pay Review Bodies (PRBs): Motivation PRB Remit Key Features of PRB Activity PM- Gordon Brown @ TUC Conference - Brighton 10th...
  • Modifiers - Auburn University

    Modifiers - Auburn University

    Modifiers are adjectives, adverbs and modifying phrases. The red car. The happy puppy. The quick fox. The feasibility report. The rapidly bouncing ball. Modifiers "modify" the meaning of the subject, verb, or complement by . describing. and . limiting.
  • IDM Chapter 104 Utility Coordination Gail Lee Utility

    IDM Chapter 104 Utility Coordination Gail Lee Utility

    RFC. for at least one utility that exists within the project limits. Exceptions: Go to . [email protected] . The undersigned affirms that they are an INDOT Certified Utility Coordinator and have performed all of the required duties for the project....
  • Mobility Studies Lauren Kark Introduction Outcome Measures Locomotor

    Mobility Studies Lauren Kark Introduction Outcome Measures Locomotor

    Lauren Kark . Introduction . Outcome Measures . Locomotor Capabilities Index Barthel Index Functional Independence Measure . Office of Population Consensus and Surveys Scale. Amputee Activity Score. Functional Measure for Amputees . Houghton Scale.
  • An Ideal Healthy Environment for Learning Using Technology A ...

    An Ideal Healthy Environment for Learning Using Technology A ...

    An Ideal Healthy Environment for Learning Using Technology A case study for Development and Continuous Education Center- Baghdad University Bahaa I. Kazem (PhD) Muntaha A K. Jasim (MSc) Development and Continuous Education Center Baghdad University- Iraq * Lifelong learning (on...
  • PNV 1914 - nemcc.edu

    PNV 1914 - nemcc.edu

    Self-Evaluation. Practical Nursingstudents took the Self-Assessment Survey on.Results were: Majority were Reflective Critical Thinkers. Critical Thinking is a construct for problem solving that is described as rational, self-regulatory, decision-making within the context of clinical judgment.
  • Classroom observations of literacy practice

    Classroom observations of literacy practice

    Woolf Fisher Research Centre. The University of Auckland. Classroom Observations Can…. Support reflective teaching practice. Introduce 'another pair of eyes' to the classroom (it can be hard to see or judge your own teaching objectively)