=Paper=
{{Paper
|id=Vol-1924/ialatecml_paper6
|storemode=property
|title=Multi-Arm Active Transfer Learning for Telugu Sentiment Analysis
|pdfUrl=https://ceur-ws.org/Vol-1924/ialatecml_paper6.pdf
|volume=Vol-1924
|authors=Subba Reddy Oota ,Vijaysaradhi Indurthi,Mounika Marreddy,Sandeep Sricharan Mukku,Radhika Mamidi
|dblpUrl=https://dblp.org/rec/conf/pkdd/OotaIMMM17
}}
==Multi-Arm Active Transfer Learning for Telugu Sentiment Analysis==
https://ceur-ws.org/Vol-1924/ialatecml_paper6.pdf
Multi-Arm Active Transfer Learning for Telugu
Sentiment Analysis
Subba Reddy Oota1 , Vijaysaradhi Indurthi1 , Mounika Marreddy2
Sandeep Sricharan Mukku1 , and Radhika Mamidi1
1
International Institute of Information Technology, Hyderabad,
2
Quadratyx, Hyderabad.
oota.subba@students.iiit.ac.in,vijaya.saradhi@research.iiit.ac.in
mounika0559@gmail.com,sandeep.mukku@research.iiit.ac.in
radhika.mamidi@iiit.ac.in
Abstract. Transfer learning algorithms can be used when sufficient
amount of training data is available in the source domain and limited
training data is available in the target domain. The transfer of knowledge
from one domain to another requires similarity between two domains. In
many resource-poor languages, it is rare to find labeled training data in
both the source and target domains. Active learning algorithms, which
query more labels from an oracle, can be used effectively in training the
source domain when an oracle is available in the source domain but not
available in the target domain. Active learning strategies are subjective
as they are designed by humans. It can be time consuming to design a
strategy and it can vary from one human to other. To tackle all these
problems, we design a learning algorithm that connects transfer learning
and active learning with the well-known multi-armed bandit problem by
querying the most valuable information from the source domain.
The advantage of our method is that we get the best active query se-
lection using active learning with multi arm and distribution matching
between two domains in conjunction with transfer learning. The effec-
tiveness of the proposed method is validated by running experiments on
three Telugu language domain-specific datasets for sentiment analysis.
Keywords: Active Learning, Transfer Learning, Multi-Arm Bandit
1 Introduction
People comment on online reviews and blog posts in social media about trend-
ing activities in their regional languages. There are many tools, resources and
corpora available to analyze these activities for English language. However, not
many tools and resources are available to analyze these activities in resource poor
languages like Telugu. With the dearth of sufficient annotated sentiment data in
the Telugu language, we need to increase the existing available labeled datasets
in different domains. However, annotating abundant unlabeled data manually is
very time-consuming, cost-ineffective, and resource-intensive.
To address the above problems, we propose a Multi-Arm Active Transfer
Learning (MATL) algorithm, which involves transfer learning [1] and a combi-
nation of query selection strategies in active learning [3]. One of the prerequisites
62
Multi-Arm
2 Active Active
Multi-Arm Transfer Learning
Transfer for Telugu Sentiment Analysis
Learning
for transfer learning is that the source and target domains should be closely re-
lated. We use Maximum Mean Discrepancy (MMD) [2] as a measure to find
the closeness between two distributions of the source and target domains. In
this paper, we experiment with sentiment analysis of Telugu language domain
specific datasets: Movies, Political and Sports1 . By considering each domain as
the source or target domain, we have a total of 6 domain pairs: M-P, M-S, P-M,
P-S, S-M, S-P. Figure 1 shows two domain pair results. We evaluate the accu-
racy with three different classification techniques viz., support vector machines
(SVM), extreme gradient boosting (XGBoost), gradient boosted trees (GBT),
and meta learning of all these approaches and record the accuracy.
2 Approach & Results
In Multi-Arm active transfer learning approach, it takes both source domain:
S = {unlabeled data instances (SU ), labeled data instances (SL )}, and target
domain: T = {unlabeled data instances (TU ), labeled data instances (TL ), test
data instances (TT ) (used for measuring classification accuracy at each itera-
tion)}, iterations (n) as an input. A decision making model is built along with
this approach to predict the posterior probability for each instance of SU . After
calculating the sampling query distribution φ(S(n)), based on multi-arm bandit
approach a best sample instance xin ∈ S is selected for querying. If xin ∈ SU ,
then this selected sample instance (xin ) is labeled with an oracle/labeler as yin
and added to SL . Now the classifier (Cn ) is trained on the total set {updated
SL ,TL }. Using MMD [2], the distance between two distributions is calculated.
This process is repeated until reached query budget. The classification model Cn
is tested on target test data TT to measure the accuracy. The reward (rn (ak (n)))
and observation(on (ak (n))) is updated by comparing the label yin given by the
oracle/labeler with the classifier (Cn (xin )).
0.72
0.7
0.7
0.68
0.65
Accuracy (%)
Accuracy (%)
0.66
0.64
0.6
0.62
0.55
Uncertainty Sampling 0.6 Uncertainty Sampling
Random Sampling Random Sampling
QUIRE 0.58
QUIRE
0.5 QBC QBC
DWUS DWUS
0.56
MATL MATL
0 50 100 150 200 250 300 350 400 450 500 0 50 100 150 200 250 300 350 400 450 500
Number of queried instances Number of queried instances
(a) P-S (b) S-P
Fig. 1. Performance comparison on Sentiment Analysis
References
1. Gong, B.: Discriminatively learning domain-invariant features for unsupervised do-
main adaptation. (2013)
2. Gretton, A., Smola, A.J.: A kernel method for the two-sample-problem (2007)
3. Settles, B.: Active learning literature survey. Tech. rep. (2010)
1
https://github.com/subbareddy248/Datasets/tree/master
63