The key innovation of the model lies in its ability to detect patterns indicative of depression, and then map those patterns to new individuals, with no additional information. “We call it ‘context-free,’ because we are not putting any constraints into the types of questions you’re looking for and the type of responses to those questions,”

Detecting depression

The researchers trained and tested their model on a dataset of 142 interactions from the Distress Analysis Interview Corpus that contains audio, text, and video interviews of patients with mental-health issues and virtual agents controlled by humans

The researchers, on the other hand, used a technique called sequence modeling, often used for speech processing. With this technique, they fed the model sequences of text and audio data from questions and answers, from both depressed and non-depressed individuals, one by one. As the sequences accumulated, the model extracted speech patterns that emerged for people with or without depression. Words such as, say, “sad,” “low,” or “down,” may be paired with audio signals that are flatter and more monotone. Individuals with depression may also speak slower and use longer pauses between words. These text and audio identifiers for mental distress have been explored in previous research. It was ultimately up to the model to determine if any patterns were predictive of depression or not.

“The model sees sequences of words or speaking style, and determines that these patterns are more likely to be seen in people who are depressed or not depressed,

She said we can meet at a particular time and agreed not knowing about the consequences that could happen .But during these meeting lots of misunderstanding happened.I am wandering if with only small meeting this can happen then how can those people who are living for long time can handle this situation.

We started our communication by India’s most popular social networking app whats app. Initially we started communicating like friends but when my friends knew this, they told me that you are thinking she is your girlfriend but she might already have boyfriend ,ask her.Then I told ,I have seen her with a boy but i have doubt whether he is her ex or not .They said ask her.On their suggestion I asked her.

What I found was that It was true.But only one question confused me “Do you like me?.I had no answer for that.

But I hid my feeling and said yes somewhat.Then i said i don’t like you but actually it was not true.

First proposal-I knew that what i am going to write might trigger her anger but we know proposing a girl is not a crime.I said ,I want to be something more than a friend .

Misunderstanding-I wanted to discuss something regarding project but She was not cooperating ,She thought he might be in love that’s why he is calling me.But the mistake might me on my side also because She might be thinking if She will not do the project alone then sir might not give marks.

DISASTER-I sent her one mail requesting her to cooperate with me.But English language has some ambiguities ,cooperate first meaning- act in accordance with a wish or command and second meaning — work jointly towards the same end.Since I have not given her any command before so the second meaning applies.I have also written don’t worry about about marks and there is chance that our research might get published.I will talk to sir regarding this .

What she thought-She considered the first meaning of cooperate ,Second she thought I am bribing her by saying I will talk to sir regarding this.

What was my intention-I said cooperate in the context of second meaning as i said i had not given her prior command.I said I will talk to sir regarding this.Here the word “this” refer’s to how to go ahead for publication and not about marks.I never tried to increase my marks then how can i tell to increase her marks, it is unethical.

What i did -I called to tell this but instead she scolded me and said never call me again .I do not want to talk you .I knew what kind of person you are .

Conclusion-Not everyone knows everything but everyone knows something but when we try to exchange knowledge this can happen.

With the help of this visual analytics government can know the various factors of death and can improve its policies.

Here is the packed bubbles of number of people died due to pregnancy for respective years

Next bubbles shows number of people died due to pregnancy who has married in particular year

The following charts shows where the people died in respective years

The following charts shows factors affecting death of people for the respective years

The following image will show in which state the more number of people died due to pregnancy

Correlation matrix

rural=1(if it is rural) ,2(if it is urban) ,deceased sex=1(male) ,2(female),is_death_registered=1(yes),2(no),3(don’t know).

An analysis of women mortality data of India. was originally published in DataDrivenInvestor on Medium, where people are continuing the conversation by highlighting and responding to this story.

Before moving further we need to understand the following layers as well as algorithms.

N-gram Convolutional Layer-This layer is a standard convolutional layer which extracts n-gram features at different positions of a sentence through various convolutional filters.

Primary Capsule Layer- This is the first capsule layer in which the capsules replace the scalar-output feature detectors of CNNs with vector-output capsules to preserve the instantiate parameters such as the local order of words and semantic representations of words.

Convolutional Capsule Layer- In this layer, each capsule is connected only to a local region spatially in the layer below. Those capsules in the region multiply transformation matrices to learn child-parent relationships followed by routing by agreement to produce parent capsules in the layer above.

Fully Connected Capsule Layer -The capsules in the layer below are flattened into a list of capsules and fed into fully connected capsule layer in which capsules are multiplied by transformation matrix followed by routing-by-agreement to produce final capsule and its probability for each category.

Dynamic Routing-The basic idea of dynamic routing is to design a non-linear map. The non-linear map is constructed in an iterative manner ensuring the output of each capsule gets sent to an appropriate parent in the subsequent layer. For each potential parent, the capsule network can increase or decrease the connection strength by dynamic routing, which is more effective than the primitive routing strategies such as max-pooling in CNN that essentially detects whether a feature is present in any position of the text, but loses spatial information about the feature. They explored three strategies to boost the accuracy of routing process by alleviating the disturbance of some noisy capsules: Orphan Category- an additional “orphan” category is added to the network, which can capture the “background” information of the text such as stop words and the words that are unrelated to specific categories, helping the capsule network model the child-parent relationship more efficiently. Adding “orphan” category in the text is more effective than in image since there is no single consistent “background” object in images, while the stop words are consistent in texts such as predicate and pronoun words. Leaky-Softmax-They explored Leaky-Softmax in the place of standard softmax while updating connection strength between the children capsules and their parents. Despite the orphan category in the last capsule layer, we also need a light-weighted method between two consecutive layers to route the noise child capsules to extra dimension without any additional parameters and computation consuming. Coefficients Amendment- They attemped to use the probability of existence of child capsules in the layer below to iteratively amend the connection strength.

Conclusion

Extensive experiments on six text classification benchmarks showed the effectiveness of capsule networks in text classification. More importantly, capsule networks also show significant improvement when transferring single-label to multi-label text classifications over strong baseline methods.

From paper titled “ Investigating Capsule Networks with Dynamic Routing for Text Classification” which was published on 20 june 2018.