=Paper=
{{Paper
|id=Vol-1329/paper1
|storemode=property
|title=Adapting Sentiment Lexicons using Contextual Semantics for Twitter Sentiment Analysis
|pdfUrl=https://ceur-ws.org/Vol-1329/paper_2.pdf
|volume=Vol-1329
}}
==Adapting Sentiment Lexicons using Contextual Semantics for Twitter Sentiment Analysis==
https://ceur-ws.org/Vol-1329/paper_2.pdf
Adapting Sentiment Lexicons using Contextual
Semantics for Sentiment Analysis of Twitter
Hassan Saif,1 Yulan He,2 Miriam Fernandez1 and Harith Alani1
1
Knowledge Media Institute, The Open University, United Kingdom
{h.saif, m.fernandez, h.alani}@open.ac.uk
2
School of Engineering and Applied Science, Aston University, United Kingdom
y.he@cantab.net
Abstract. Sentiment lexicons for sentiment analysis offer a simple, yet effective
way to obtain the prior sentiment information of opinionated words in texts.
However, words’ sentiment orientations and strengths often change throughout
various contexts in which the words appear. In this paper, we propose a lexicon
adaptation approach that uses the contextual semantics of words to capture their
contexts in tweet messages and update their prior sentiment orientations and/or
strengths accordingly. We evaluate our approach on one state-of-the-art sentiment
lexicon using three different Twitter datasets. Results show that the sentiment
lexicons adapted by our approach outperform the original lexicon in accuracy and
F-measure in two datasets, but give similar accuracy and slightly lower F-measure
in one dataset.
Keywords: Sentiment Analysis, Semantics, Lexicon Adaptation, Twitter
1 Introduction
Sentiment analysis on Twitter has been attracting much attention recently due to the
rapid growth in Twitter’s popularity as a platform for people to express their opinions
and attitudes towards a great variety of topics. Most existing approaches to Twitter
sentiment analysis can be categorised into machine learning [7, 11, 13] and lexicon-
based approaches [2, 8, 15, 6].
Lexicon-based approaches use lexicons of words weighted with their sentiment
orientations to determine the overall sentiment in texts. These approaches have shown
to be more applicable to Twitter data than machine learning approaches, since they do
not require training from labelled data and therefore, they offer a domain-independent
sentiment detection [15]. Nonetheless, lexicon-based approaches are limited by the
sentiment lexicon used [21]. Firstly, because sentiment lexicons are composed by a
generally static set of words that do not cover the wide variety of new terms that
constantly emerge in the social web. Secondly, because words in the lexicons have fixed
prior sentiment orientations, i.e. each term has always the same associated sentiment
orientation independently of the context in which the term is used.
To overcome the above limitations, several lexicon bootstrapping and adaptation
methods have been previously proposed. However, these methods are either supervised
[16], i.e., they require training from human-coded corpora, or based on studying the
statistical, syntactical or linguistic relations between words in general textual corpora
(e.g., The Web) [17, 19] or in static lexical knowledge sources (e.g., WordNet) [5]
6
ignoring, therefore, the specific textual context in which the words appear. In many
cases, however, the sentiment of a word is implicitly associated with the semantics of its
context [3].
In this paper we propose an unsupervised approach for adapting sentiment lexicons
based on the contextual semantics of their words in a tweet corpus. In particular, our
approach studies the co-occurrences between words to capture their contexts in tweets
and update their prior sentiment orientations and/or sentiment strengths in a given lexicon
accordingly.
As a case study we apply our approach on Thelwall-Lexicon [15], which, to our
knowledge, is the state-of-the-art sentiment lexicon for social data. We evaluate the
adapted lexicons by performing a lexicon-based polarity sentiment detection (positive vs.
negative) on three Twitter datasets. Our results show that the adapted lexicons produce
a significant improvement in the sentiment detection accuracy and F-measure in two
datasets but gives a slightly lower F-measure in one dataset.
In the rest of this paper, related work is discussed in Section 2, and our approach is
presented in Sections 3. Experiments and results are presented in Sections 4. Discussion
and future work are covered in Section 5. Finally, we conclude our work in Section 6.
2 Related Work
Exiting approaches to bootstrapping and adapting sentiment lexicons can be categorised
into dictionary and corpus-based approaches. The dictionary-based approach [5, 14] starts
with a small set of general opinionated words (e.g., good, bad) and lexical knowledge base
(e.g., WordNet). After that, the approach expands this set by searching the knowledge
base for words that have lexical or linguistic relations to the opinionated words in the
initial set (e.g., synonyms, glosses, etc).
Alternatively, the corpus-based approach measures the sentiment orientation of
words automatically based on their association to other strongly opinionated words in a
given corpus [17, 14, 19]. For example, Turney and Littman [17] used Pointwise Mutual
Information (PMI) to measure the statistical correlation between a given word and a
balanced set of 14 positive and negative paradigm words (e.g., good, nice, nasty, poor).
Although this work does not require large lexical input knowledge, its identification
speed is very limited [21] because it uses web search engines in order to retrieve the
relative co-occurrences of words.
Following the aforementioned approaches, several lexicons such as MPQA [20]
and SentiWordNet [1] have been induced and successfully used for sentiment analysis
on conventional text (e.g., movie review data). However, on Twitter these lexicons are
not as compatible due to their limited coverage of Twitter-specific expressions, such as
abbreviations and colloquial words (e.g, “looov”, “luv”, “gr8”) that are often found
in tweets.
Quite few sentiment lexicons have been recently built to work specifically with social
media data, such as Thelwall-Lexicon [16] and Nielsen-Lexicon [8]. These lexicons have
proven to work effectively on Twitter data. Nevertheless, such lexicons are similar to
other traditional ones, in the sense that they all offer fixed and context-insensitive word-
sentiment orientations and strengths. Although a training algorithm has been proposed
to update the sentiment of terms in Thelwall-Lexicon[16], it requires to be trained from
human-coded corpora, which is labour-intensive to obtain.
7
Aiming at addressing the above limitations we have designed our lexicon-adaptation
approach in away that allows to (i) work in unsupervised fashion, avoiding the need for
labelled data, and (ii) exploit the contextual semantics of words. This allows capturing
their contextual information in tweets and update their prior sentiment orientation and
strength in a given sentiment lexicon accordingly.
3 A Contextual Semantic Approach to Lexicon Adaptation
The main principle behind our approach is that the senti-
ment of a term is not static, as found in general-purpose Tweets
Extract
sentiment lexicons, but rather depends on the context in Contextual
Sentiment Sentiment
which the term is used, i.e., it depends on its contextual Lexicon
semantics.3 Therefore, our approach functions in two main
steps as shown in Figure 1. First, given a tweet collection Rule-based Lexicon Adaptation
and a sentiment lexicon, the approach builds a contextual
semantic representation for each unique term in the tweet
Adapted Lexicon
collection and subsequently uses it to derive the term’s con-
textual sentiment orientation and strength. The SentiCircle Fig. 1. The systematic work-
representation model is used to this end [10]. Secondly, flow of our proposed lexicon
rule-based algorithm is applied to amend the prior senti- adaptation approach.
ment of terms in the lexicon based on their corresponding
contextual sentiment. Both steps are further detailed in the following subsections.
3.1 Capturing Contextual Semantics and Sentiment
The first step in our pipeline is to capture the words contextual semantics and sentiment
in tweets. To this end, we use our previously proposed semantic representation model,
SentiCircle [10].
Following the distributional hypothesis that words that co-occur in similar contexts
tend to have similar meaning [18], SentiCircle extracts the contextual semantics of
a word from its co-occurrence patterns with other words in a given tweet collection.
These patterns are then represented as a geometric circle, which is subsequently used
to compute the contextual sentiment of the word by applying simple trigonometric
identities on it. In particular, for each unique term m in a tweet collection, we build
a two-dimensional geometric circle, where the term m is situated in the centre of the
circle, and each point around it represents a context term ci (i.e., a term that occurs with
m in the same context). The position of ci , as illustrated in Figure 2, is defined jointly
by its Cartesian coordinates xi , yi as:
xi = ri cos(✓i ⇤ ⇡) yi = ri sin(✓i ⇤ ⇡)
Where ✓i is the polar angle of the context term ci and its value equals to the prior
sentiment of ci in a sentiment lexicon before adaptation, ri is the radius of ci and its
value represents the degree of correlation (tdoc) between ci and m, and can be computed
as:
N
ri = tdoc(m, ci ) = f (ci , m) ⇥ log
N ci
3
We define context as a textual corpus or a set of tweets.
8
where f (ci , m) is the number of times ci occurs with m in tweets, N is the total
number of terms, and Nci is the total number of terms that occur with ci . Note that
all terms’ radii in the SentiCircle are normalised. Also, all angles’ values are in radian.
The trigonometric properties of the SentiCircle allows us Y
+1
to encode the contextual semantics of a term as sentiment Very Positive Positive
orientation and sentiment strength. Y-axis defines the sen- y i
C i
timent of the term, i.e., a positive y value denotes a positive r i
θ
sentiment and vice versa. The X-axis defines the sentiment -1 +1
i
X
m x
strength of the term. The smaller the x value, the stronger
i
the sentiment.4 This, in turn, divides the circle into four sen-
timent quadrants. Terms in the two upper quadrants have
Very Negative Negative
a positive sentiment (sin ✓ > 0), with upper left quadrant -1
representing stronger positive sentiment since it has larger r = TDOC(C )
i
i
i
θ = Prior_Sentiment (C ) i
angle values than those in the top right quadrant. Simi- Fig. 2. SentiCircle of a term
larly, terms in the two lower quadrants have negative sen- m. Neutral region is shaded
timent values (sin ✓ < 0). Moreover, a small region called in blue.
the “Neutral Region” can be defined. This region, as shown
in Figure 2, is located very close to X-axis in the “Positive” and the “Negative” quadrants
only, where terms lie in this region have very weak sentiment (i.e, |✓| t 0).
Calculating Contextual Sentiment In summary, the Senti-Circle of a term m is com-
posed by the set of (x, y) Cartesian coordinates of all the context terms of m. An effective
way to compute the overall sentiment of m is by calculating the geometric median of all
the points in its SentiCircle. Formally, for a given set of n points (p1 , p2 ,P
..., pn ) in a Senti-
n
Cirlce ⌦, the 2D geometric median g is defined as: g = arg ming2R2 i=1 k|pi g||2 .
We call the geometric median g the SentiMedian as its position in the SentiCircle
determines the final contextual-sentiment orientation and strength of m.
Note that the boundaries of the neutral region can be computed by measuring the
density distribution of terms in the SentiCircle along the Y-axis. In this paper we use
similar boundaries to the ones used in [10] since we use the same evaluation datasets.
3.2 Lexicon Adaptation
The second step in our approach is to update the sentiment lexicon with the terms’
contextual sentiment information extracted in the previous step. As mentioned earlier, in
this work we use Thelwall-Lexicon [16] as a case study. Therefore, in this section we
first describe this lexicon and its properties, and then introduce our proposed adaptation
method.
Thelwall-Lexicon consists of 2546 terms coupled with integer values between -5 (very
negative) and +5 (very positive). Based on the terms’ prior sentiment orientations and
strengths (SOS), we group them into three subsets of 1919 negative terms (SOS2[-2,-5]),
398 positive terms (SOS2[2,5]) and 229 neutral terms (SOS2{-1,1}).
The adaptation method uses a set of antecedent-consequent rules that decides how the
prior sentiment of the terms in Thelwall-Lexicon should be updated according to the
positions of their SentiMedians (i.e., their contextual sentiment). In particular, for a term
m, the method checks (i) its prior SOS value in Thelwall-Lexicon and (ii) the SentiCircle
4
This is because cos ✓ < 0 for large angles.
9
quadrant in which the SentiMedian of m resides. The method subsequently chooses the
best-matching rule to update the term’s prior sentiment and/or strength.
Table 1 shows the complete list of rules in the proposed method. As noted, these rules
are divided into updating rules, i.e., rules for updating the existing terms in Thelwall-
Lexicon, and expanding rules, i.e., rules for expanding the lexicon with new terms. The
updating rules are further divided into rules that deal with terms that have similar prior
and contextual sentiment orientations (i.e., both positive or negative), and rules that deal
with terms that have different prior and contextual sentiment orientations (i.e., negative
prior, positive contextual sentiment and vice versa).
Although they look complicated, the notion behind the proposed rules is rather simple:
Check how strong the contextual sentiment is and how weak the prior sentiment is !
update the sentiment orientation and strength accordingly. The strength of the contextual
sentiment can be determined based on the sentiment quadrant of the SentiMedian of m,
i.e., the contextual sentiment is strong if the SentiMedian resides in the “Very Positive”
or “Very Negative” quadrants (See Figure 2). On the other hand, the prior sentiment of
m (i.e., priorm ) in Thelwall-Lexicon is weak if |priorm | 6 3 and strong otherwise.
Updating Rules (Similar Sentiment Orientations)
Id Antecedents Consequent
1 (|prior| 6 3) ^ (SentiM edian 2 / StrongQuadrant) |prior| = |prior| + 1
2 (|prior| 6 3) ^ (SentiM edian 2 StrongQuadrant) |prior| = |prior| + 2
3 (|prior| > 3) ^ (SentiM edian 2 / StrongQuadrant) |prior| = |prior| + 1
4 (|prior| > 3) ^ (SentiM edian 2 StrongQuadrant) |prior| = |prior| + 1
Updating Rules (Different Sentiment Orientations)
5 (|prior| 6 3) ^ (SentiM edian 2 / StrongQuadrant) |prior| = 1
6 (|prior| 6 3) ^ (SentiM edian 2 StrongQuadrant) prior = prior
7 (|prior| > 3) ^ (SentiM edian 2 / StrongQuadrant) |prior| = |prior| 1
8 (|prior| > 3) ^ (SentiM edian 2 StrongQuadrant) prior = prior
9 (|prior| > 3) ^ (SentiM edian 2 N eutralRegion) |prior| = |prior| 1
10 (|prior| 6 3) ^ (SentiM edian 2 N eutralRegion) |prior| = 1
Expanding Rules
11 SentiM edian 2 N eutralRegion (|contextual| = 1) ^ AddT erm
12 SentiM edian 2 / StrongQuadrant (|contextual| = 3) ^ AddT erm
13 SentiM edian 2 StrongQuadrant (|contextual| = 5) ^ AddT erm
Table 1. Adaptation rules for Thelwall-Lexicon, where prior: prior sentiment value,
StrongQuadrant: very negative/positive quadrant in the SentiCircle, Add: add the term to
Thelwall-Lexicon.
For example, the word “revolution” in Thelwall-Lexicon has a weak negative
sentiment (prior=-2) while it has a neutral contextual sentiment since its SentiMedian re-
sides in the neutral region (SentiM edian 2 N eutralRegion). Therefore, rule number
10 is applied and the term’s prior sentiment in Thelwall lexicon will be updated to neutral
(|prior| = 1). In another example, the words “Obama” and “Independence” are not
covered by the Thelwall-Lexicon, and therefore, they have no prior sentiment. However,
their SentiMedians reside in the “Positive” quadrant in their SentiCircles, and therefore
rule number 12 is applied and both terms will be assigned with a positive sentiment
strength of 3 and added to the lexicon consequently.
4 Evaluation Results
We evaluate our approach on Thelwall-Lexicon using three adaptation settings: (i) the
update setting where we update the prior sentiment of existing terms in the lexicon, (ii)
The expand setting where we expand Thelwall-Lexicon with new opinionated terms, and
(iii) the update+expand setting where we try both aforementioned settings together. To
10
this end, we use three Twitter datasets OMD, HCR and STS-Gold. Numbers of positive
and negative tweets within these datasets are summarised in Table 2, and detailed in the
references added in the table. To evaluate the adapted lexicons under the above settings,
we perform binary polarity classification on the three datasets. To this end, we use the
sentiment detection method proposed with Thelwall-Lexicon [15]. According to this
method a tweet is considered as positive if its aggregated positive sentiment strength is
1.5 times higher than the aggregated negative one, and negative vice versa.
Dataset Tweets Positive Negative
Obama-McCain Debate (OMD)[4] 1081 393 688
Health Care Reform (HCR)[12] 1354 397 957
Standford Sentiment Gold Standard (STS-Gold)[9] 2034 632 1402
Table 2. Twitter datasets used for the evaluation
Applying our adaptation approach to Thelwall-Lexicon results in dramatic changes
in it. Table 3 shows the percentage of words in the three datasets that were found in
Thelwall-Lexicon with their sentiment changed after adaptation. One can notice that on
average 9.61% of the words in our datasets were found in the lexicon. However, updating
the lexicon with the contextual sentiment of words resulted in 33.82% of these words
flipping their sentiment orientation and 62.94% changing their sentiment strength while
keeping their prior sentiment orientation. Only 3.24% of the words in Thelwall-Lexicon
remained untouched. Moreover, 21.37% of words previously unseen in the lexicon were
assigned with contextual sentiment by our approach and added to Thelwall-Lexicon
subsequently.
OMD HCR STS-Gold Average
Words found in the lexicon 12.43 8.33 8.09 9.61
Hidden words 87.57 91.67 91.91 90.39
Words flipped their sentiment orientation 35.02 35.61 30.83 33.82
Words changed their sentiment strength 61.83 61.95 65.05 62.94
Words remained unchanged 3.15 2.44 4.13 3.24
New opinionated words 23.94 14.30 25.87 21.37
Table 3. Average percentage of words in the three datasets that had their sentiment orientation or
strength updated by our adaptation approach
Table 4 shows the average results of binary sentiment classification performed on
our datasets using (i) the original Thelwall-Lexicon (Original), (ii) Thelwall-Lexicon
induced under the update setting (Updated), and (iii) Thelwall-Lexicon induced under
the update+expand setting.5 The table reports the results in accuracy and three sets of
precision (P), recall (R), and F-measure (F1), one for positive sentiment detection, one
for negative, and one for the average of the two.
From these results in Table 4, we notice that the best classification performance in
accuracy and F1 is obtained on the STS-Gold dataset regardless the lexicon being used.
We also observe that the negative sentiment detection performance is always higher than
the positive detection performance for all datasets and lexicons.
As for different lexicons, we notice that on OMD and STS-Gold the adapted lexicons
outperform the original lexicon in both accuracy and F-measure. For example, on OMD
the adapted lexicon shows an average improvement of 2.46% and 4.51% in accuracy and
F1 respectively over the original lexicon. On STS-Gold the performance improvement is
5
Note that in this work we do not report the results obtained under the expand setting since no
improvement was observed comparing to the other two settings.
11
Positive Sentiment Negative Sentiment Average
Datasets Lexicons Accuracy
P R F1 P R F1 P R F1
Original 66.79 55.99 40.46 46.97 70.64 81.83 75.82 63.31 61.14 61.4
OMD Updated 69.29 58.89 51.4 54.89 74.12 79.51 76.72 66.51 65.45 65.8
Updated+Expanded 69.2 58.38 53.18 55.66 74.55 78.34 76.4 66.47 65.76 66.03
Original 66.99 43.39 41.31 42.32 76.13 77.64 76.88 59.76 59.47 59.6
HCR Updated 67.21 42.9 35.77 39.01 75.07 80.25 77.58 58.99 58.01 58.29
Updated+Expanded 66.99 42.56 36.02 39.02 75.05 79.83 77.37 58.8 57.93 58.19
Original 81.32 68.75 73.1 70.86 87.52 85.02 86.25 78.13 79.06 78.56
STS-Gold Updated 81.71 69.46 73.42 71.38 87.7 85.45 86.56 78.58 79.43 78.97
Updated+Expanded 82.3 70.48 74.05 72.22 88.03 86.02 87.01 79.26 80.04 79.62
Table 4. Cross comparison results of original and the adapted lexicons
less significant than that on OMD, but we still observe 1% improvement in accuracy and
F1 comparing to using the original lexicon. As for the HCR dataset, the adapted lexicon
gives on average similar accuracy, but 1.36% lower F-measure. This performance drop
can be attributable to the poor detection performance of positive tweets. Specifically,
we notice from Table 4 a major loss in the recall on positive tweet detection using both
adapted lexicons. One possible reason is the sentiment class distribution in our datasets.
In particular, one may notice that HCR is the most imbalanced amongst the three datasets.
Moreover, by examining the numbers in Table 3, we can see that HCR presents the lowest
number of new opinionated words among the three datasets (i.e., 10.61% lower than the
average) which could be another potential reason for not observing any performance
improvement.
5 Discussion and Future Work
We demonstrated the value of using contextual semantics of words for adapting senti-
ment lexicons from tweets. Specifically, we used Thelwall-Lexicon as a case study and
evaluated its adaptation to three datasets of different sizes. Although the potential is
palpable, our results were not conclusive, where a performance drop was observed in the
HCR dataset using our adapted lexicons. Our initial observations suggest that the quality
of our approach might be dependent on the sentiment class distribution in the dataset.
Therefore, a deeper investigation in this direction is required.
We used the SentiCircle approach to extract the contextual semantics of words from
tweets. In future work we will try other contextual semantic approaches and study how
the semantic extraction quality affects the adaptation performance.
Our adaptation rules in this paper are specific to Thelwall-Lexicon. These rules,
however, can be generalized to other lexicons, which constitutes another future direction
of this work.
All words which have contextual sentiment were used for adaptation. Nevertheless,
the results conveyed that the prior sentiments in the lexicon might need to be unchanged
for words of specific syntactical or linguistic properties in tweets. Part of our future work
is to detect and filter those words that are more likely to have stable sentiment regardless
the contexts in which they appear.
6 Conclusions
In this paper we proposed an unsupervised approach for sentiment lexicon adapta-
tion from Twitter data. Our approach extracts the contextual semantics of words and
uses them to update the words’ prior sentiment orientations and/or strength in a given
sentiment lexicon. The evaluation was done on Thelwall-Lexicon using three Twitter
datasets. Results showed that lexicons adapted by our approach improved the sentiment
classification performance in both accuracy and F1 in two out of three datasets.
12
Acknowledgment
This work was supported by the EU-FP7 project SENSE4US (grant no. 611242).
References
1. Baccianella, S., Esuli, A., Sebastiani, F.: Sentiwordnet 3.0: An enhanced lexical resource for
sentiment analysis and opinion mining. In: Seventh conference on International Language
Resources and Evaluation, Malta. Retrieved May. Valletta, Malta (2010)
2. Bollen, J., Mao, H., Zeng, X.: Twitter mood predicts the stock market. Journal of Computa-
tional Science 2(1), 1–8 (2011)
3. Cambria, E.: An introduction to concept-level sentiment analysis. In: Advances in Soft
Computing and Its Applications, pp. 478–483. Springer (2013)
4. Diakopoulos, N., Shamma, D.: Characterizing debate performance via aggregated twitter
sentiment. In: Proc. 28th Int. Conf. on Human factors in computing systems. ACM (2010)
5. Esuli, A., Sebastiani, F.: Determining term subjectivity and term orientation for opinion
mining. In: EACL. vol. 6, p. 2006 (2006)
6. Hu, X., Tang, J., Gao, H., Liu, H.: Unsupervised sentiment analysis with emotional signals.
In: Proceedings of the 22nd World Wide Web conf (2013)
7. Kouloumpis, E., Wilson, T., Moore, J.: Twitter sentiment analysis: The good the bad and the
omg! In: Proceedings of the ICWSM. Barcelona, Spain (2011)
8. Nielsen, F.Å.: A new anew: Evaluation of a word list for sentiment analysis in microblogs.
arXiv preprint arXiv:1103.2903 (2011)
9. Saif, H., Fernandez, M., He, Y., Alani, H.: Evaluation datasets for twitter sentiment analysis a
survey and a new dataset, the sts-gold. In: Proceedings, 1st ESSEM Workshop. Turin, Italy
(2013)
10. Saif, H., Fernandez, M., He, Y., Alani, H.: Senticircles for contextual and conceptual semantic
sentiment analysis of twitter. In: Proc. 11th Extended Semantic Web Conf. (ESWC). Crete,
Greece (2014)
11. Saif, H., He, Y., Alani, H.: Semantic sentiment analysis of twitter. In: Proc. 11th Int. Semantic
Web Conf. (ISWC). Boston, MA (2012)
12. Speriosu, M., Sudan, N., Upadhyay, S., Baldridge, J.: Twitter polarity classification with label
propagation over lexical links and the follower graph. In: Proceedings of the EMNLP First
workshop on Unsupervised Learning in NLP. Edinburgh, Scotland (2011)
13. Suttles, J., Ide, N.: Distant supervision for emotion classification with discrete binary values.
In: Computational Linguistics and Intelligent Text Processing, pp. 121–136. Springer (2013)
14. Takamura, H., Inui, T., Okumura, M.: Extracting semantic orientations of words using spin
model. In: Proc. 43rd Annual Meeting on Association for Computational Linguistics (2005)
15. Thelwall, M., Buckley, K., Paltoglou, G.: Sentiment strength detection for the social web. J.
American Society for Information Science and Technology 63(1), 163–173 (2012)
16. Thelwall, M., Buckley, K., Paltoglou, G., Cai, D., Kappas, A.: Sentiment strength detection in
short informal text. J. American Society for Info. Science and Technology 61(12) (2010)
17. Turney, P., Littman, M.: Measuring praise and criticism: Inference of semantic orientation
from association. ACM Transactions on Information Systems 21, 315–346 (2003)
18. Turney, P.D., Pantel, P., et al.: From frequency to meaning: Vector space models of semantics.
Journal of artificial intelligence research 37(1), 141–188 (2010)
19. Velikovich, L., Blair-Goldensohn, S., Hannan, K., McDonald, R.: The viability of web-derived
polarity lexicons. In: Human Language Technologies: ACL (2010)
20. Wilson, T., Wiebe, J., Hoffmann, P.: Recognizing contextual polarity in phrase-level sentiment
analysis. In: Proc. Empirical Methods in NLP Conf. (EMNLP). Vancouver, Canada (2005)
21. Xu, T., Peng, Q., Cheng, Y.: Identifying the semantic orientation of terms using s-hal for
sentiment analysis. Knowledge-Based Systems 35, 279–289 (2012)