Figure 5 - uploaded by Ilona Södervik
Content may be subject to copyright.
Similarity comparisons of lexicons. The similarity S JSD based on Jensen-Shannon divergence is shown in the upper row for contingency C (at left) and concurrence Q (middle). The changes ∆S due to tuning are also shown (right). All values are shown as deviations from the median (value provided above bar-legend). The cosine similarity S COS is shown in the lower row. In both cases, results are for a stabilized state with β = 50.
Source publication
Complex networks are often used to analyze written text and reports by rendering texts in the form of a semantic network, forming a lexicon of words or key terms. Many existing methods to construct lexicons are based on counting word co-occurrences, having the advantage of simplicity and ease of applicability. Here, we use a quantum semantics appro...
Contexts in source publication
Context 1
... cosine similarity S COS in Equation (8), on the other hand, is based on the correlation centrality of nodes, i.e., off-diagonal sums of the correlation matrix. Figure 5 shows the similarities S JSD and S COS (upper and lower panels, respectively) for C-and Q-based lexicons (left and middle panels, respectively) as deviations from the median similarity (values shown above legends). The absolute change ∆S in similarity when C-based lexicons are optimized to Q-based lexicons is shown in Figure 5 in the right-hand panels. ...
Context 2
... 5 shows the similarities S JSD and S COS (upper and lower panels, respectively) for C-and Q-based lexicons (left and middle panels, respectively) as deviations from the median similarity (values shown above legends). The absolute change ∆S in similarity when C-based lexicons are optimized to Q-based lexicons is shown in Figure 5 in the right-hand panels. Results in Figure 5 for S JSD show that in all cases the increase in similarity due to optimization is significant, although relative patterns as related to median values are somewhat similar, as already seen in three examples in Figure 4. ...
Context 3
... absolute change ∆S in similarity when C-based lexicons are optimized to Q-based lexicons is shown in Figure 5 in the right-hand panels. Results in Figure 5 for S JSD show that in all cases the increase in similarity due to optimization is significant, although relative patterns as related to median values are somewhat similar, as already seen in three examples in Figure 4. However, some interesting changes in relative patterns are notable. ...
Context 4
... A1, on the other hand, is most similar to A2, OC, and V2, in that order. Contrary to these more or less expected findings, Figure 5 shows that sets D1, D2, and D3 of lexicons are less similar to each other than they are to lexicon A1; the authors of D1, D2, and D3 have to some degree different vocabularies in their texts, but the review article manages to cover all of them. In addition, V1, V2, and V3 all have quite low similarity to D1, D2, and D3, which was expected, because the main dividing line of thought is known to be between these disciplinary sub-groups. ...
Context 5
... the results in Figures 5 and 6, we can conclude that some lexicons of some texts are highly overlapping. For example, lexicons V1 and V2 are quite similar to each other, as are also lexicons P1 and P2 and A1 and A2. ...
Context 6
... a comparison can be carried out using JSD-similarity, which can be calculated from Equations (6) and (7) for ordinary probability distributions by replacing density matrices with probability density distributions and traces with sums. The resulting pairwise similarities of lexicons are similar to results shown in Figures 5 and 6, and again the same high similarity clusters, as discussed above, can be discerned. For comparison, the Kendall-tau correlations of frequency-based lexicon JSD-similarities to concurrence and contingency-based similarities are now 0.37 and 0.35, respectively. ...
Context 7
... is noteworthy that in both C-and Q-lexicons, the effect of β on the rankings of the upper-most terms was not very significant. This is in agreement with the result that cos-similarity (which weights key terms) did not change dramatically with increasing value of β (see Figures 5 and 6). ...
Similar publications
Feature extraction of electrical load plays a vital role in providing a reliable basis and guidance for power companies. In this paper, we propose a novel clustering algorithm named the Density-based Matrix Transformation (DBMT) Clustering method to extract features (peaks, valleys and trends) of electrical load curves. The main objective of the al...
Citations
... Such strong disciplinary fragmentation seems to be particularly apparent and typical in the human and behavioral sciences [23][24][25]. Another situation of interest, where the creation and generation of new knowledge is not necessarily of primary interest, but where differing disciplinary views about thematic topics can be recognized, is related to the disciplinary views of science education scholars [26,27] as well as science students, where student groups may have consensus views that differ from those of other student groups views, even when they have used the same study materials [28,29]. To address situations in which the knowledge or meaning structures of interest are complex systems of terms, concepts, or conceptions characteristic of the disciplinary group, it seems appropriate to use the expression "webs of beliefs" (compare e.g., ref. [30]), to be referred to briefly as WoBs in what follows. ...
Formation of consensus groups with shared opinions or views is a common feature of human social life and also a well-known phenomenon in cases when views are complex, as in the case of the formation of scholarly disciplines. In such cases, shared views are not simple sets of opinions but rather complex webs of beliefs (WoBs). Here, we approach such consensus group formation through the agent-based model (ABM). Agents’ views are described as complex, extensive web-like structures resembling semantic networks, i.e., webs of beliefs. In the ABM introduced here, the agents’ interactions and participation in sharing their views are dependent on the similarity of the agents’ webs of beliefs; the greater the similarity, the more likely the interaction and sharing of elements of WoBs. In interactions, the WoBs are altered when agents seek consensus and consensus groups are formed. The consensus group formation depends on the agents’ sensitivity to the similarity of their WoBs. If their sensitivity is low, only one large and diffuse group is formed, while with high sensitivity, many separated and segregated consensus groups emerge. To conclude, we discuss how such results resemble the formation of disciplinary, scholarly consensus groups.
